Self-adaptive cache dynamic adjustment method

By monitoring the operating status and port traffic information of the switch in real time and adjusting the cache strategy dynamically, the problem that traditional cache management strategies cannot adapt to load changes is solved, and the cache utilization efficiency is improved.

CN118890328AActive Publication Date: 2024-11-01INSPUR SUZHOU INTELLIGENT TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411139428.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-01
Estimated Expiration
2044-08-19

AI Technical Summary

Technical Problem

Traditional static cache management strategies cannot adapt to the switch's real-time load and data access mode changes, resulting in inefficiency in cache and waste of resources.

Method used

By monitoring the switch's operating status and port traffic information, the load and cache usage data are calculated in real time, and the cache adjustment strategy is dynamically adjusted by generating cache adjustment strategies compared with preset benchmarks.

Benefits of technology

Maximize cache utilization efficiency, adapt to changes in system load and data access mode, and improve the operation efficiency of the switch.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118890328A_ABST
    Figure CN118890328A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides a self-adaptive cache dynamic adjustment method, which comprises the following steps of: monitoring the running state of a switch, periodically acquiring port flow information of the switch, calculating load data of the switch in a working state according to the port flow information, comparing the load data with a preset load performance reference, and judging whether the load data is in the working state or not. Calculating current cache use information of the switch according to the port flow information, and comparing the cache use information with a preset cache performance reference to obtain a second evaluation result for the switch; and finally, combining the first evaluation result and the second evaluation result to generate a cache adjustment strategy for the switch, and adjusting the cache of the switch according to the cache adjustment strategy. According to the embodiment of the invention, the cache capacity and the replacement strategy are automatically adjusted by monitoring the real-time working load and the cache utilization condition of the switch, so that the utilization efficiency of the cache is maximized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of network communication, and in particular to an adaptive cache dynamic adjustment method. Background Art

[0002] With the rapid development of network technology, cache technology plays a vital role. By storing data in the cache, the number of accesses to the main memory can be reduced, thereby increasing the data processing speed. However, as the core device in the network, the switch needs to process more and more data as the amount of data continues to grow and the access frequency continues to increase, and the data access pattern is becoming more and more complex. Traditional static cache management strategies can no longer meet the requirements of efficient operation of switches. In the prior art, cache management in switches mainly relies on fixed cache size and replacement strategies, or by collecting historical data, using machine learning algorithms to train prediction models, and then guiding cache replacement and updates based on the prediction results.

[0003] The current cache replacement strategy is fixed and cannot adapt to changes in system load and data access patterns, resulting in low cache efficiency and resource waste. It cannot be dynamically adjusted according to the real-time load and data access patterns of the switch. Summary of the invention

[0004] The embodiment of the present invention provides an adaptive cache dynamic adjustment method, which realizes real-time monitoring of the system's operating status and data access mode, and dynamically adjusts the cache strategy using an intelligent algorithm to adapt to changes in system load and data access mode. The specific technical solution is as follows:

[0005] In a first aspect of the present invention, a method for adaptive cache dynamic adjustment is first provided, the method comprising:

[0006] Monitor the operating status of the switch and regularly collect port traffic information of the switch;

[0007] Calculating load data of the switch in a working state according to the port traffic information, and comparing the load data with a preset load performance benchmark to obtain a first evaluation result for the switch, wherein the first evaluation result is used to indicate a working load status of the switch;

[0008] Calculating the current cache usage information of the switch according to the port traffic information, and comparing the cache usage information with a preset cache performance benchmark to obtain a second evaluation result for the switch, wherein the second evaluation result is used to indicate the cache utilization degree of the switch;

[0009] The first evaluation result and the second evaluation result are combined to generate a cache adjustment policy for the switch, and the cache of the switch is adjusted according to the cache adjustment policy.

[0010] Optionally, the method further includes:

[0011] Monitoring a new operating state of the switch after applying the cache adjustment policy and generating a corresponding log record, wherein the log record includes an application time point of the cache adjustment policy, new load data after the switch is adjusted, and new cache usage information;

[0012] The log records are analyzed periodically, and a new cache adjustment policy for the switch is generated according to the log records.

[0013] Optionally, the switch includes a plurality of ports, and the monitoring of the operating status of the switch and the periodic collection of port traffic information of the switch include:

[0014] Collecting real-time traffic data of the port through a traffic counter, and determining the traffic category of the port according to the real-time traffic data, wherein the traffic category includes a high-traffic port, a low-traffic port, and an abnormal-traffic port;

[0015] Regularly query the accumulated flow data collected by the flow counter within a fixed time, and calculate the flow change of the port within a unit time according to the accumulated flow data;

[0016] Acquire flow information of the real-time flow data, the flow information including a source address, a destination address and a port number of the real-time flow data, and analyze the flow information to obtain an access mode corresponding to the real-time flow data;

[0017] Monitoring the total number of data requests passing through the switch within a fixed time, the data request being used to request a data item in the cache, the total number including the number of successful requests and the number of failed requests;

[0018] When the data item corresponding to the data request exists in the cache, recording the status of the data request as a successful request;

[0019] When the data item corresponding to the data request does not exist in the cache, the status of the data request is recorded as request failure.

[0020] Optionally, the load data includes average flow data, frame loss rate and delay time, and the calculation of the load data of the switch in the working state according to the port flow information includes:

[0021] Calculate the average flow data per unit time according to the port flow information;

[0022] Collecting frame data transmitted by the switch in a continuous working state through a frame counter, the frame data including the number of input frames, the number of output frames and the number of discarded frames, and calculating a frame loss rate of the switch when transmitting the frame data according to the number of input frames and the number of discarded frames;

[0023] Recording an input timestamp corresponding to the input frame number and an output timestamp corresponding to the output frame number;

[0024] The difference between the input timestamp and the output timestamp is calculated to obtain the delay time of the switch when transmitting the frame data.

[0025] Optionally, comparing the load data with a preset load performance benchmark includes:

[0026] When the load data exceeds the load performance benchmark, the switch is judged to be in a high load state; when the load data is within the load performance benchmark, the switch is judged to be in a normal load state; when the load data is less than the load performance benchmark, the switch is judged to be in a low load state;

[0027] Wherein, when the load data exceeds the load performance benchmark, judging that the switch is in a high load state includes:

[0028] Presetting a load performance benchmark of the switch in a continuous working state, the load performance benchmark including an average flow benchmark, a frame loss rate benchmark and a delay time benchmark;

[0029] The average flow data is compared with the average flow benchmark to obtain a first load performance comparison result of the switch, and according to the first load performance comparison result, the switch whose average flow data exceeds the average flow benchmark is regarded as a high-load state switch;

[0030] Comparing the frame loss rate with the frame loss rate benchmark to obtain a second load performance comparison result of the switch, and according to the second load performance comparison result, determining the switch whose frame loss rate exceeds the frame loss rate benchmark as a high-load state switch;

[0031] The delay time is compared with the delay time benchmark to obtain a third load performance comparison result of the switch, and the switch whose delay time exceeds the delay time benchmark is regarded as a high-load state switch according to the third load performance comparison result.

[0032] Optionally, the cache usage information includes a cache hit rate and a cache usage rate, the cache includes a plurality of cache layers, and calculating the current cache usage information of the switch according to the port traffic information, and comparing the cache usage information with a preset cache performance benchmark, includes:

[0033] Calculate the cache hit rate of the cache layer according to the total number of times and the number of successful requests;

[0034] Calculating the cache utilization rate of the cache layer according to the total space capacity of the cache and the occupied space capacity of the data item;

[0035] Presetting a cache performance benchmark of the switch in a normal working state, the cache performance benchmark including a cache hit rate benchmark and a cache utilization rate benchmark;

[0036] Comparing the cache hit rate with the cache hit rate benchmark, judging that the cache layer is in a high cache hit rate state when the cache hit rate exceeds the cache hit rate benchmark, judging that the cache layer is in a normal cache hit rate state when the cache hit rate is within the cache hit rate benchmark, and judging that the cache layer is in a low cache hit rate state when the cache hit rate is less than the cache hit rate benchmark;

[0037] The cache utilization is compared with the cache utilization benchmark. When the cache utilization exceeds the cache utilization benchmark, the cache layer is judged to be in a high cache utilization state. When the cache utilization is within the cache utilization benchmark, the cache layer is judged to be in a normal cache utilization state. When the cache utilization is less than the cache utilization benchmark, the cache layer is judged to be in a low cache utilization state.

[0038] Optionally, combining the first evaluation result and the second evaluation result to generate a cache adjustment policy for the switch, and adjusting the cache of the switch according to the cache adjustment policy includes:

[0039] When the switch is in the high load state and the cache layer is in the low cache hit rate, the expansion space capacity required by the switch is calculated in combination with the load performance benchmark and the cache hit rate benchmark, and the total space capacity of the cache is increased according to the expansion space capacity;

[0040] When the switch is in the low load state or the cache layer is in the low cache utilization state, a reduced space capacity for the switch is calculated according to the load performance benchmark or the cache utilization benchmark, and a total space capacity of the cache is reduced according to the reduced space capacity;

[0041] When the switch is in the high cache utilization state, data items in the cache are deleted until the cache utilization is within the cache utilization benchmark.

[0042] In a second aspect of the present invention, there is also provided an adaptive cache dynamic adjustment device, the device comprising:

[0043] A port traffic information collection module is used to monitor the operating status of the switch and regularly collect port traffic information of the switch;

[0044] a first evaluation result generating module, configured to calculate the load data of the switch in a working state according to the port flow information, and compare the load data with a preset load performance benchmark to obtain a first evaluation result for the switch, wherein the first evaluation result is used to indicate a working load status of the switch;

[0045] a second evaluation result generating module, configured to calculate the current cache usage information of the switch according to the port traffic information, and compare the cache usage information with a preset cache performance benchmark to obtain a second evaluation result for the switch, wherein the second evaluation result is used to indicate the cache utilization degree of the switch;

[0046] The cache adjustment strategy generating module is used to combine the first evaluation result and the second evaluation result to generate a cache adjustment strategy for the switch, and adjust the cache of the switch according to the cache adjustment strategy.

[0047] In a third aspect of the implementation of the present invention, there is also provided a communication device, comprising: a transceiver, a memory, a processor, and a program stored in the memory and executable on the processor;

[0048] The processor is used to read the program in the memory to implement any of the above-mentioned adaptive cache dynamic adjustment methods.

[0049] In a fourth aspect of the implementation of the present invention, a computer-readable storage medium is further provided, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer executes any of the above-mentioned adaptive cache dynamic adjustment methods.

[0050] The adaptive cache dynamic adjustment method provided by the embodiment of the present invention monitors the running state of the switch and regularly collects the port flow information of the switch, calculates the load data of the switch in the working state according to the port flow information, and compares the load data with the preset load performance benchmark to obtain a first evaluation result for the switch, and the first evaluation result is used to indicate the workload status of the switch; calculates the current cache usage information of the switch according to the port flow information, and compares the cache usage information with the preset cache performance benchmark to obtain a second evaluation result for the switch, and the second evaluation result is used to indicate the cache utilization degree of the switch; finally, the first evaluation result and the second evaluation result are combined to generate a cache adjustment strategy for the switch, and the cache of the switch is adjusted according to the cache adjustment strategy. The embodiment of the present invention automatically adjusts the cache capacity and replacement strategy by monitoring the real-time workload and cache utilization of the switch to maximize the utilization efficiency of the cache. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0052] Figure 1 A flowchart of the steps of the adaptive cache dynamic adjustment method provided by an embodiment of the present invention;

[0053] Figure 2 It is a structural schematic diagram of an adaptive cache dynamic adjustment device provided by an embodiment of the present invention;

[0054] Figure 3 It is a structural diagram of a communication device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it can be understood by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable readers to better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present invention can also be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other without contradiction.

[0056] Reference Figure 1 , shows a flowchart of the steps of the adaptive cache dynamic adjustment method provided by an embodiment of the present invention, the method may include:

[0057] Step 101, monitor the operating status of the switch and regularly collect port traffic information of the switch.

[0058] The switch in the embodiment of the present invention includes several ports. The collected port traffic information includes real-time traffic data of the port, cumulative traffic data of the port within a fixed time, access mode corresponding to the real-time traffic data, and the total number of data requests passing through the switch within a fixed time. The specific implementation steps are as follows:

[0059] The real-time traffic data of the port is collected by the traffic counter, and the traffic category of the port is determined according to the real-time traffic data. The traffic categories include high-traffic ports, low-traffic ports and abnormal traffic ports. Real-time traffic data refers to the amount of data being transmitted on the switch port at a specific moment. For example, at a certain moment (such as 10:00:00), the real-time traffic data of the port is 100Mbps (Megabits per second), indicating that at this moment, the port is transmitting data at a speed of 100Mbps. In one embodiment, the maximum value of real-time traffic and the minimum value of real-time traffic can be set. When the real-time traffic threshold collected by the traffic counter is greater than or equal to the maximum value of real-time traffic, the port is determined to be a high-traffic port, and when the real-time traffic threshold collected by the traffic counter is less than the minimum value of real-time traffic, the port is determined to be a low-traffic port. For abnormal traffic ports, the historical traffic data of the port can be combined for analysis and determination. When the real-time traffic data of the port increases sharply or decreases sharply compared with the historical traffic data, the port is determined to be an abnormal traffic port. In another embodiment, if a port is in a high-traffic state for a period of time, it can also be determined to be an abnormal traffic port.

[0060] Through the above method, the traffic status of the port can be identified, so that for high-traffic ports, their cache allocation can be increased when specifying the cache adjustment strategy to ensure timely processing and forwarding of data. For low-traffic ports, their cache allocation can be appropriately reduced to release more cache resources for high-traffic ports. When cache resources are tight, a priority strategy can be adopted to give priority to the cache requirements of important data flows or services.

[0061] The traffic counter is periodically queried for the accumulated traffic data collected within a fixed time, and the traffic change of the port within a unit time is calculated based on the accumulated traffic data. The accumulated traffic data refers to the total amount of data transmitted on the switch port from a certain starting time point to the current time point. In one embodiment, a fixed time interval for querying the traffic counter can be determined, such as every minute, every hour, or every day, and the accumulated traffic data of the switch port within the fixed time interval is periodically queried, and the collected accumulated traffic data is recorded. Between two query time points, the difference in accumulated traffic data is calculated to obtain the traffic change of the port during this period. For example:

[0062] The starting time is: January 1, 2024 0:00:00

[0063] First query time: January 1, 2024 10:00:00

[0064] Second query time: January 1, 2024 11:00:00

[0065] Cumulative traffic data for the first query: 100,000 bytes

[0066] Cumulative traffic data for the second query: 150,000 bytes

[0067] Among them, flow change = cumulative flow data of the second query - cumulative flow data of the first query

[0068] Therefore, the traffic change this time = 150,000 bytes - 100,000 bytes = 50,000 bytes

[0069] The flow information of the real-time flow data is obtained, the flow information includes the source address, the destination address and the port number of the real-time flow data, and the flow information is analyzed to obtain the access mode corresponding to the real-time flow data. In one example, port mirroring can be configured on the switch to mirror the port flow to be monitored to an analysis port, thereby capturing the flow data packets of the real-time data flow on the analysis port, extracting the source address, the destination address and the port number from the captured flow data packets, analyzing the combination of the source address, the destination address and the port number, and identifying the access mode corresponding to the flow data packet.

[0070] Monitor the total number of data requests that pass through the switch within a fixed time. The data request is used to request data items in the cache. The total number includes the number of successful requests and the number of failed requests. When the data item corresponding to the data request exists in the cache, the status of the data request is recorded as a successful request. When the data item corresponding to the data request does not exist in the cache, the status of the data request is recorded as a failed request.

[0071] Step 102, calculate the load data of the switch in the working state according to the port traffic information, and compare the load data with the preset load performance benchmark to obtain a first evaluation result for the switch, and the first evaluation result is used to indicate the working load status of the switch. In one embodiment, the load data includes average traffic data, frame loss rate and delay time, and the specific execution steps include:

[0072] Calculate the average traffic data per unit time based on the port traffic information. Preset the time interval, add all the real-time traffic data collected within the time interval to get the total traffic, divide the total traffic by the time interval to get the average traffic data. For example, if the time interval is set to 1 hour (60 minutes), and the total traffic collected within 1 hour is 60,000 bytes, then the corresponding average traffic data is 60,000 bytes / hour, or 1000 bytes / minute.

[0073] The frame counter collects frame data transmitted by the switch in a continuous working state. The frame data includes the number of input frames, the number of output frames, and the number of discarded frames. The frame loss rate of the switch when transmitting frame data is calculated based on the number of input frames and the number of discarded frames.

[0074] Record the input timestamp corresponding to the input frame number and the output timestamp corresponding to the output frame number.

[0075] The difference between the input timestamp and the output timestamp is calculated to obtain the delay time of the switch when transmitting the frame data.

[0076] When the load data exceeds the load performance benchmark, the switch is judged to be in a high load state; when the load data is within the load performance benchmark, the switch is judged to be in a normal load state; when the load data is less than the load performance benchmark, the switch is judged to be in a low load state.

[0077] When the load data exceeds the load performance benchmark, it is determined that the switch is in a high load state, including:

[0078] The load performance benchmark of the switch in a continuous working state is preset, and the load performance benchmark includes an average flow benchmark, a frame loss rate benchmark and a delay time benchmark.

[0079] The average flow data is compared with the average flow benchmark to obtain a first load performance comparison result of the switch, and according to the first load performance comparison result, the switch whose average flow data exceeds the average flow benchmark is regarded as a high-load state switch. The frame loss rate is compared with the frame loss rate benchmark to obtain a second load performance comparison result of the switch, and according to the second load performance comparison result, the switch whose frame loss rate exceeds the frame loss rate benchmark is regarded as a high-load state switch. The delay time is compared with the delay time benchmark to obtain a third load performance comparison result of the switch, and according to the third load performance comparison result, the switch whose delay time exceeds the delay time benchmark is regarded as a high-load state switch.

[0080] Step 103, calculate the current cache usage information of the switch according to the port traffic information, and compare the cache usage information with the preset cache performance benchmark to obtain a second evaluation result for the switch, and the second evaluation result is used to indicate the cache utilization of the switch. In an embodiment of the present invention, the cache usage information includes a cache hit rate and a cache usage rate, and the cache includes several cache layers. The specific implementation steps include:

[0081] The cache hit rate of the cache layer is calculated based on the total number of times and the number of successful requests.

[0082] The cache utilization of the cache layer is calculated based on the total space capacity of the cache and the space capacity occupied by the data items.

[0083] The cache performance benchmark of the switch under normal working conditions is preset, and the cache performance benchmark includes a cache hit rate benchmark and a cache utilization rate benchmark.

[0084] The cache hit rate is compared with the cache hit rate benchmark. When the cache hit rate exceeds the cache hit rate benchmark, the cache layer is judged to be in a high cache hit rate state. When the cache hit rate is within the cache hit rate benchmark, the cache layer is judged to be in a normal cache hit rate state. When the cache hit rate is less than the cache hit rate benchmark, the cache layer is judged to be in a low cache hit rate state.

[0085] The cache utilization is compared with the cache utilization benchmark. When the cache utilization exceeds the cache utilization benchmark, the cache layer is judged to be in a high cache utilization state. When the cache utilization is within the cache utilization benchmark, the cache layer is judged to be in a normal cache utilization state. When the cache utilization is less than the cache utilization benchmark, the cache layer is judged to be in a low cache utilization state.

[0086] Through the above method, the cache hit rate and utilization status can be identified, so that the data can be allocated to different cache layers according to its importance. For data with high importance, it is allocated to the high-speed cache layer with a high cache hit rate state. For data with average importance, it is allocated to the normal-speed cache layer with a normal cache hit rate state. For data with low importance, it is allocated to the low-speed cache layer with a low cache hit rate state.

[0087] Step 104: Combine the first evaluation result and the second evaluation result to generate a cache adjustment strategy for the switch, and adjust the cache of the switch according to the cache adjustment strategy. The specific implementation steps include:

[0088] When the switch is in a high load state and the cache layer is in a low cache hit rate, the expansion space capacity required by the switch is calculated in combination with the load performance benchmark and the cache hit rate benchmark, and the total space capacity of the cache is increased according to the expansion space capacity.

[0089] When the switch is in a low load state or the cache layer is in a low cache utilization state, the reduced space capacity for the switch is calculated according to the load performance benchmark or the cache utilization benchmark, and the total space capacity of the cache is reduced according to the reduced space capacity.

[0090] When the cache layer of the switch is in a high cache utilization state, data items in the cache are deleted until the cache utilization is within a cache utilization benchmark.

[0091] The adaptive cache dynamic adjustment method provided by the embodiment of the present invention monitors the running state of the switch and regularly collects the port flow information of the switch, calculates the load data of the switch in the working state according to the port flow information, and compares the load data with the preset load performance benchmark to obtain a first evaluation result for the switch, and the first evaluation result is used to indicate the workload status of the switch; calculates the current cache usage information of the switch according to the port flow information, and compares the cache usage information with the preset cache performance benchmark to obtain a second evaluation result for the switch, and the second evaluation result is used to indicate the cache utilization degree of the switch; finally, the first evaluation result and the second evaluation result are combined to generate a cache adjustment strategy for the switch, and the cache of the switch is adjusted according to the cache adjustment strategy. The embodiment of the present invention automatically adjusts the cache capacity and replacement strategy by monitoring the real-time workload and cache utilization of the switch to maximize the utilization efficiency of the cache.

[0092] The adaptive cache dynamic adjustment method provided by the embodiment of the present invention also includes the following steps:

[0093] Monitor the new running state of the switch after applying the cache adjustment policy and generate corresponding log records, the log records include the application time point of the cache adjustment policy, the new load data of the switch after adjustment, and the new cache usage information;

[0094] The log records are analyzed periodically and new cache tuning policies for the switch are generated based on the log records.

[0095] In an embodiment of the present invention, firstly, the current cache strategy is evaluated based on the data collected by the monitoring system, such as cache hit rate, traffic data fluctuation, etc. If it is found that the cache hit rate decreases or some service requests are not effectively processed, it is necessary to adjust the cache capacity or optimize the parameter settings of the replacement strategy to improve the cache efficiency. Secondly, the adjusted cache strategy is redeployed to the switch and its execution effect is continuously monitored. This includes real-time monitoring of key indicators such as cache usage, hit rate, latency, and collecting user feedback and service performance data. Then, according to the new monitoring and evaluation results, the strategy is optimized again. This process may involve adjusting the cache size, updating the replacement algorithm, optimizing the cache preheating mechanism, etc. Through continuous iterative optimization, a cache management mechanism that dynamically adapts to network changes and service needs is formed. In addition, in order to ensure the effect and security of cache adjustment, relevant information and suggestions on cache adjustment can be provided to developers through log records, alarm notifications, etc. At the same time, cache adjustment is combined with other network optimization technologies, such as routing optimization and traffic scheduling, to achieve cross-layer network performance optimization and improve the stability and efficiency of the overall network.

[0096] In the embodiment of the present invention, for the identified abnormal traffic port, a reset isolation operation can be performed on it, and the specific implementation steps are as follows:

[0097] Analyze the traffic data of the abnormal traffic port. The traffic data may be real-time traffic data or cumulative traffic data, and the present invention is not limited to this. According to the characteristics of the traffic data, identify the corresponding abnormal pattern. The abnormal pattern may include a sudden increase in traffic, a change in traffic pattern, and continuous high traffic.

[0098] Then adjust the network configuration of the abnormal traffic port according to its abnormal mode and re-monitor it. The number of times to adjust the network configuration and re-monitor can be set. If abnormal traffic still exists after multiple readjustments, the abnormal traffic port will be isolated.

[0099] In one embodiment of the present invention, the cache replacement strategy can also be automatically adjusted according to the most recent use time and access frequency of the cached data items. The specific implementation steps are as follows:

[0100] LRU algorithm (Least Recently Used): Based on the principle of most recent usage time, cache items that have not been used for the longest time are eliminated.

[0101] LFU (Least Frequently Used) algorithm: Based on the principle of frequency of use, the cache items with the lowest frequency of use are eliminated.

[0102] Record the most recent access time and access frequency of each data item in the cache. Specifically, a timestamp can be recorded for each data item to indicate the last time the cache item was accessed, and a data structure (such as a linked list, queue, hash table, etc.) can be used to store and manage data items and their access times. Whenever a data item is accessed, its timestamp is updated to the current time.

[0103] In the LRU algorithm, a linked list can be used to maintain the access order of data items. Whenever a data item is accessed, it is moved to the head of the linked list, indicating that it has been recently accessed. At the same time, a hash table is used to quickly find the position of the data item in the linked list so that the access time can be quickly updated.

[0104] In the LFU algorithm, in addition to recording the access time, it is also necessary to record the number of accesses to each data item. Use frequency buckets (such as bucket sorting) to manage data items with different access frequencies so that the data item with the lowest access frequency can be quickly found.

[0105] The most recent access time and access frequency analyze the cache usage pattern of each data item and identify hot data and cold data. You can set the time base value and frequency base value. Hot data refers to data items whose timestamps are after the time base value or whose access frequency is not less than the frequency base value. Cold data refers to data items whose timestamps are before the time base value or whose access frequency is less than the frequency base value.

[0106] Based on the analysis results, the cache replacement strategy is dynamically adjusted and implemented, so that hot data can be retained in the cache to improve response speed, and cold data can be cleared from the cache to free up space.

[0107] In the embodiment of the present invention, communication with the switch and policy application can be achieved through the API interface or SDK provided by the switch, and the storage location of the data can be dynamically adjusted according to specific needs. For example, the API can be used to specify that certain data items should always be stored in the cache layer, or the cache policy can be dynamically adjusted according to the state of the application. Since large data items will occupy a large amount of memory resources and may cause other important data items to be eliminated, it can be specified that large data items are stored separately in a fixed cache layer, and important data items may require more reliable storage and faster access speeds. Therefore, it can be specified that data items with higher importance are stored in the cache layer.

[0108] In the embodiment of the present invention, the corresponding validity period of each data item in the cache can also be set according to its size. Specifically, the maxmemory parameter can be set in the Redis configuration file to limit the maximum size of the cache and select an appropriate replacement strategy. For example, setting maxmemory 1gb means that the maximum size of the cache is 1GB, and the replacement strategy may include:

[0109] volatile-lru: Use the LRU algorithm to eliminate data items with expiration times.

[0110] allkeys-lru: Use the LRU algorithm to eliminate all data items.

[0111] volatile-lfu: Use the LFU algorithm to eliminate data items with expiration times.

[0112] allkeys-lfu: Use the LFU algorithm to eliminate all data items.

[0113] volatile-random: Randomly eliminate data items with expiration times set.

[0114] allkeys-random: randomly eliminate all data items.

[0115] volatile-ttl: Eliminate the data item with an expiration time set and the shortest remaining lifetime.

[0116] noeviction: Do not evict data items and return an error when the cache is full.

[0117] Through the above method, the two adaptive adjustment modes of time and access frequency can be effectively combined to better balance the cache size and the validity period of data items, thereby improving the performance and efficiency of the cache.

[0118] Reference Figure 2 , shows a schematic diagram of the structure of an adaptive cache dynamic adjustment device provided by an embodiment of the present invention, such as Figure 2 As shown, the device may include:

[0119] The port traffic information collection module 201 is used to monitor the operation status of the switch and regularly collect the port traffic information of the switch.

[0120] The first evaluation result generating module 202 is used to calculate the load data of the switch in the working state according to the port traffic information, and compare the load data with the preset load performance benchmark to obtain the first evaluation result for the switch, and the first evaluation result is used to indicate the working load status of the switch.

[0121] The second evaluation result generating module 203 is used to calculate the current cache usage information of the switch according to the port traffic information, and compare the cache usage information with a preset cache performance benchmark to obtain a second evaluation result for the switch, and the second evaluation result is used to indicate the cache utilization degree of the switch.

[0122] The cache adjustment strategy generating module 204 is used to combine the first evaluation result and the second evaluation result to generate a cache adjustment strategy for the switch, and adjust the cache of the switch according to the cache adjustment strategy.

[0123] Optionally, the adaptive cache dynamic adjustment device further includes:

[0124] The log recording module is used to monitor the new operating state of the switch after applying the cache adjustment policy and generate corresponding log records, which include the application time point of the cache adjustment policy, the new load data after the switch is adjusted, and the new cache usage information.

[0125] Feedback optimization module, which is used to periodically analyze log records and generate new cache adjustment policies for switches based on the log records.

[0126] Optionally, the switch includes a plurality of ports, and the port traffic information collection module 201 further includes:

[0127] The real-time traffic data collection module is used to collect the real-time traffic data of the port through the traffic counter, and judge the traffic category of the port according to the real-time traffic data. The traffic categories include high-traffic ports, low-traffic ports and abnormal traffic ports.

[0128] The cumulative flow data query module is used to periodically query the cumulative flow data collected by the flow counter within a fixed time, and calculate the flow change of the port within a unit time based on the cumulative flow data.

[0129] The flow information analysis module is used to obtain the flow information of the real-time flow data. The flow information includes the source address, the destination address and the port number of the real-time flow data. The flow information is analyzed to obtain the access mode corresponding to the real-time flow data.

[0130] A total number monitoring module is used to monitor the total number of data requests passing through the switch within a fixed time. The data request is used to request data items in the cache. The total number includes the number of successful requests and the number of failed requests.

[0131] A request success recording module is used to record the status of the data request as a successful request when there is a data item corresponding to the data request in the cache;

[0132] The request failure recording module is used to record the status of the data request as request failure when there is no data item corresponding to the data request in the cache.

[0133] Optionally, the load data includes average flow data, frame loss rate and delay time, and the first evaluation result generating module 202 further includes:

[0134] The average flow data calculation module is used to calculate the average flow data per unit time according to the port flow information.

[0135] The frame loss rate calculation module is used to collect frame data transmitted by the switch in a continuous working state through a frame counter. The frame data includes the number of input frames, the number of output frames and the number of discarded frames, and calculate the frame loss rate of the switch when transmitting frame data based on the number of input frames and the number of discarded frames.

[0136] The timestamp recording module is used to record the input timestamp corresponding to the input frame number and the output timestamp corresponding to the output frame number.

[0137] The delay time calculation module is used to calculate the difference between the input timestamp and the output timestamp to obtain the delay time of the switch when transmitting frame data.

[0138] Optionally, the first evaluation result generating module 202 further includes:

[0139] A load status judgment module, used to judge that the switch is in a high load state when the load data exceeds the load performance benchmark, judge that the switch is in a normal load state when the load data is within the load performance benchmark, and judge that the switch is in a low load state when the load data is less than the load performance benchmark;

[0140] Wherein, the load status judgment module includes:

[0141] A load performance benchmark setting module is used to pre-set the load performance benchmark of the switch in a continuous working state. The load performance benchmark includes an average flow benchmark, a frame loss rate benchmark and a delay time benchmark;

[0142] A first high-load state switch determination module is used to compare the average flow data with the average flow benchmark to obtain a first load performance comparison result of the switch, and according to the first load performance comparison result, determine the switch whose average flow data exceeds the average flow benchmark as a high-load state switch;

[0143] a second high-load state switch determination module, configured to compare the frame loss rate with the frame loss rate benchmark to obtain a second load performance comparison result of the switch, and to determine, according to the second load performance comparison result, a switch whose frame loss rate exceeds the frame loss rate benchmark as a high-load state switch;

[0144] The third high-load state switch determination module is used to compare the delay time with the delay time benchmark to obtain a third load performance comparison result of the switch, and according to the third load performance comparison result, the switch whose delay time exceeds the delay time benchmark is regarded as a high-load state switch.

[0145] Optionally, the cache usage information includes a cache hit rate and a cache usage rate, the cache includes several cache layers, and the second evaluation result generating module 203 further includes:

[0146] A cache hit rate calculation module is used to calculate the cache hit rate of the cache layer based on the total number of times and the number of successful requests;

[0147] A cache utilization calculation module, used to calculate the cache utilization of the cache layer according to the total space capacity of the cache and the space capacity occupied by the data items;

[0148] A cache performance benchmark setting module, used to pre-set a cache performance benchmark of the switch under normal working conditions, wherein the cache performance benchmark includes a cache hit rate benchmark and a cache utilization rate benchmark;

[0149] A cache hit rate state judgment module, used to compare the cache hit rate with a cache hit rate benchmark, to judge that the cache layer is in a high cache hit rate state when the cache hit rate exceeds the cache hit rate benchmark, to judge that the cache layer is in a normal cache hit rate state when the cache hit rate is within the cache hit rate benchmark, and to judge that the cache layer is in a low cache hit rate state when the cache hit rate is less than the cache hit rate benchmark;

[0150] The cache utilization status judgment module is used to compare the cache utilization with the cache utilization benchmark. When the cache utilization exceeds the cache utilization benchmark, it is judged that the cache layer is in a high cache utilization state; when the cache utilization is within the cache utilization benchmark, it is judged that the cache layer is in a normal cache utilization state; when the cache utilization is less than the cache utilization benchmark, it is judged that the cache layer is in a low cache utilization state.

[0151] Optionally, the cache adjustment strategy generating module 204 further includes:

[0152] The expansion module is used to calculate the expansion space capacity required by the switch in combination with the load performance benchmark and the cache hit rate benchmark when the switch is in a high load state and the cache layer is in a low cache hit rate, and increase the total space capacity of the cache according to the expansion space capacity;

[0153] A capacity reduction module, used for calculating a reduced space capacity for the switch according to a load performance benchmark or a cache utilization benchmark when the switch is in a low load state or the cache layer is in a low cache utilization state, and reducing the total space capacity of the cache according to the reduced space capacity;

[0154] The deletion module is used to delete data items in the cache when the cache layer of the switch is in a high cache utilization state until the cache utilization is within a cache utilization benchmark.

[0155] The adaptive cache dynamic adjustment method provided by the embodiment of the present invention monitors the running state of the switch and regularly collects the port flow information of the switch, calculates the load data of the switch in the working state according to the port flow information, and compares the load data with the preset load performance benchmark to obtain a first evaluation result for the switch, and the first evaluation result is used to indicate the workload status of the switch; calculates the current cache usage information of the switch according to the port flow information, and compares the cache usage information with the preset cache performance benchmark to obtain a second evaluation result for the switch, and the second evaluation result is used to indicate the cache utilization degree of the switch; finally, the first evaluation result and the second evaluation result are combined to generate a cache adjustment strategy for the switch, and the cache of the switch is adjusted according to the cache adjustment strategy. The embodiment of the present invention automatically adjusts the cache capacity and replacement strategy by monitoring the real-time workload and cache utilization of the switch to maximize the utilization efficiency of the cache.

[0156] The embodiment of the present invention also provides a communication device, such as Figure 3 As shown, it includes a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304.

[0157] Memory 303, used for storing computer programs;

[0158] The processor 301 is used to execute the program stored in the memory 303 to implement the following steps:

[0159] Monitor the operating status of the switch and regularly collect the port traffic information of the switch;

[0160] Calculating load data of the switch in a working state according to the port traffic information, and comparing the load data with a preset load performance benchmark to obtain a first evaluation result for the switch, the first evaluation result being used to indicate a working load status of the switch;

[0161] Calculating the current cache usage information of the switch according to the port traffic information, and comparing the cache usage information with a preset cache performance benchmark to obtain a second evaluation result for the switch, the second evaluation result being used to indicate the cache utilization degree of the switch;

[0162] The first evaluation result and the second evaluation result are combined to generate a cache adjustment policy for the switch, and the cache of the switch is adjusted according to the cache adjustment policy.

[0163] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0164] The communication interface is used for communication between the above terminal and other devices.

[0165] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0166] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0167] The present invention also provides a readable storage medium. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the adaptive cache dynamic adjustment method of the above-mentioned embodiment.

[0168] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0169] The algorithm and display provided herein are not inherently related to any particular computer, virtual device or other equipment. According to the above description, it is obvious that the structure required for constructing this type of device is. In addition, the present invention is not directed to any specific programming language. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description of the above specific language is to disclose the best mode of the present invention.

[0170] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0171] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than those expressly recited in each claim. Rather, as reflected in the claims below, inventive aspects lie in less than all of the features of the individual embodiments disclosed above. Therefore, the claims that follow the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present invention.

[0172] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0173] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the sorting device according to the present invention. The present invention may also be implemented as a device or apparatus program for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0174] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0175] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, apparatuses and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0176] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0177] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0178] It should be noted that the various data-related processes in the embodiments of the present invention are all carried out in compliance with the corresponding data protection laws and policies of the country where the device is located and with the authorization given by the corresponding device owner.

Claims

1. An adaptive cache dynamic adjustment method, characterized in that: The method comprises: Monitor the operating status of the switch and regularly collect port traffic information of the switch; Calculating load data of the switch in a working state according to the port traffic information, and comparing the load data with a preset load performance benchmark to obtain a first evaluation result for the switch, wherein the first evaluation result is used to indicate a working load status of the switch; Calculating the current cache usage information of the switch according to the port traffic information, and comparing the cache usage information with a preset cache performance benchmark to obtain a second evaluation result for the switch, wherein the second evaluation result is used to indicate the cache utilization degree of the switch; The first evaluation result and the second evaluation result are combined to generate a cache adjustment policy for the switch, and the cache of the switch is adjusted according to the cache adjustment policy.

2. The method according to claim 1, characterized in that The method further comprises: Monitoring a new operating state of the switch after applying the cache adjustment policy and generating a corresponding log record, wherein the log record includes an application time point of the cache adjustment policy, new load data after the switch is adjusted, and new cache usage information; The log records are analyzed periodically, and a new cache adjustment policy for the switch is generated according to the log records.

3. The method according to claim 1, characterized in that: The switch includes a plurality of ports, and the monitoring of the operating status of the switch and the regular collection of the port flow information of the switch include: Collecting real-time traffic data of the port through a traffic counter, and determining the traffic category of the port according to the real-time traffic data, wherein the traffic category includes a high-traffic port, a low-traffic port, and an abnormal-traffic port; Regularly query the accumulated flow data collected by the flow counter within a fixed time, and calculate the flow change of the port within a unit time according to the accumulated flow data; Acquire flow information of the real-time flow data, the flow information including a source address, a destination address and a port number of the real-time flow data, and analyze the flow information to obtain an access mode corresponding to the real-time flow data; Monitoring the total number of data requests passing through the switch within a fixed time, the data request being used to request a data item in the cache, the total number including the number of successful requests and the number of failed requests; When the data item corresponding to the data request exists in the cache, recording the status of the data request as a successful request; When the data item corresponding to the data request does not exist in the cache, the status of the data request is recorded as request failure.

4. The method according to claim 1, characterized in that: The load data includes average flow data, frame loss rate and delay time, and the load data of the switch in the working state is calculated according to the port flow information, including: Calculate the average flow data per unit time according to the port flow information; Collecting frame data transmitted by the switch in a continuous working state through a frame counter, the frame data including the number of input frames, the number of output frames and the number of discarded frames, and calculating a frame loss rate of the switch when transmitting the frame data according to the number of input frames and the number of discarded frames; Recording an input timestamp corresponding to the input frame number and an output timestamp corresponding to the output frame number; The difference between the input timestamp and the output timestamp is calculated to obtain the delay time of the switch when transmitting the frame data.

5. The method according to claim 3, characterized in that: The comparing the load data with a preset load performance benchmark includes: When the load data exceeds the load performance benchmark, the switch is judged to be in a high load state; when the load data is within the load performance benchmark, the switch is judged to be in a normal load state; when the load data is less than the load performance benchmark, the switch is judged to be in a low load state; Wherein, when the load data exceeds the load performance benchmark, judging that the switch is in a high load state includes: Presetting a load performance benchmark of the switch in a continuous working state, the load performance benchmark including an average flow benchmark, a frame loss rate benchmark and a delay time benchmark; The average flow data is compared with the average flow benchmark to obtain a first load performance comparison result of the switch, and according to the first load performance comparison result, the switch whose average flow data exceeds the average flow benchmark is regarded as a high-load state switch; Comparing the frame loss rate with the frame loss rate benchmark to obtain a second load performance comparison result of the switch, and according to the second load performance comparison result, determining the switch whose frame loss rate exceeds the frame loss rate benchmark as a high-load state switch; The delay time is compared with the delay time benchmark to obtain a third load performance comparison result of the switch, and the switch whose delay time exceeds the delay time benchmark is regarded as a high-load state switch according to the third load performance comparison result.

6. The method according to claim 5, characterized in that The cache usage information includes a cache hit rate and a cache usage rate, the cache includes a plurality of cache layers, and the current cache usage information of the switch is calculated according to the port traffic information, and the cache usage information is compared with a preset cache performance benchmark, including: Calculate the cache hit rate of the cache layer according to the total number of times and the number of successful requests; Calculating the cache utilization rate of the cache layer according to the total space capacity of the cache and the occupied space capacity of the data item; Presetting a cache performance benchmark of the switch in a normal working state, the cache performance benchmark including a cache hit rate benchmark and a cache utilization rate benchmark; Comparing the cache hit rate with the cache hit rate benchmark, judging that the cache layer is in a high cache hit rate state when the cache hit rate exceeds the cache hit rate benchmark, judging that the cache layer is in a normal cache hit rate state when the cache hit rate is within the cache hit rate benchmark, and judging that the cache layer is in a low cache hit rate state when the cache hit rate is less than the cache hit rate benchmark; The cache utilization is compared with the cache utilization benchmark. When the cache utilization exceeds the cache utilization benchmark, the cache layer is judged to be in a high cache utilization state. When the cache utilization is within the cache utilization benchmark, the cache layer is judged to be in a normal cache utilization state. When the cache utilization is less than the cache utilization benchmark, the cache layer is judged to be in a low cache utilization state.

7. The method according to claim 6, characterized in that The combining the first evaluation result and the second evaluation result to generate a cache adjustment strategy for the switch, and adjusting the cache of the switch according to the cache adjustment strategy, includes: When the switch is in the high load state and the cache layer is in the low cache hit rate, the expansion space capacity required by the switch is calculated in combination with the load performance benchmark and the cache hit rate benchmark, and the total space capacity of the cache is increased according to the expansion space capacity; When the switch is in the low load state or the cache layer is in the low cache utilization state, a reduced space capacity for the switch is calculated according to the load performance benchmark or the cache utilization benchmark, and a total space capacity of the cache is reduced according to the reduced space capacity; When the switch is in the high cache utilization state, data items in the cache are deleted until the cache utilization is within the cache utilization benchmark.

8. An adaptive cache dynamic adjustment device, characterized in that: The device comprises: A port traffic information collection module is used to monitor the operating status of the switch and regularly collect port traffic information of the switch; a first evaluation result generating module, configured to calculate the load data of the switch in a working state according to the port flow information, and compare the load data with a preset load performance benchmark to obtain a first evaluation result for the switch, wherein the first evaluation result is used to indicate a working load status of the switch; a second evaluation result generating module, configured to calculate the current cache usage information of the switch according to the port traffic information, and compare the cache usage information with a preset cache performance benchmark to obtain a second evaluation result for the switch, wherein the second evaluation result is used to indicate the cache utilization degree of the switch; The cache adjustment strategy generating module is used to combine the first evaluation result and the second evaluation result to generate a cache adjustment strategy for the switch, and adjust the cache of the switch according to the cache adjustment strategy.

9. A communication device, characterized in that: include: A transceiver, a memory, a processor, and a program stored on the memory and executable on the processor; The processor is used to read the program in the memory to implement the steps in the adaptive cache dynamic adjustment method as described in any one of claims 1-7.

10. A readable storage medium for storing a program, characterized in that: When the program is executed by a processor, the steps in the adaptive cache dynamic adjustment method as described in any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Network intermediate node cache optimization method based on flow characteristic analysis

    CN103023801A

  • Flow-aware switch shared cache scheduling method and device

    CN113507423A

  • Switch cache scheduling method and system

    CN114244738A

  • Data networking switch caching method, device and equipment oriented to scientific data center

    CN118488108A

  • System and method for populating a cache using behavioral adaptive policies

    US20140310373A1