Proxy Caching Engine Latency Calculation via Pipeline RTT Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining latency in communication networks, especially those using pipelined protocols like SMBv2, face significant errors due to failure to account for delays in caching engines, network equipment inactivity, command dependencies, and queuing, leading to inaccurate round trip time (RTT) and time savings (TS) calculations.
Innovation Solution
Implement a method that uses variables and flags to track delays and dependencies, such as the outstanding commands (OC) counter, inactive period (IP) flag, and other variables like FRTAIP, PRTT, LRT, URTT, and DTime, to accurately determine RTT and TS by aggregating pipeline round trip times and considering wait times and command dependencies, while employing SMBv2 protocol for pipelined command processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional RTT measurement methods are used in pipelined protocols, then the measurement process is simple, but the measurement precision deteriorates due to unaccounted delays
Solution Approach 1:
The patent segments the RTT measurement process into distinct phases: pipeline initiation phase, steady-state phase, and pipeline completion phase. Each phase is tracked using separate variables (OC counter, IP flag, FRTAIP timestamp, PRTT calculation) to accurately capture delays specific to each segment, thereby improving measurement precision without overwhelming complexity
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors pipeline state through the OC counter and IP flag, adjusting RTT calculations based on real-time observations of command dependencies and network equipment inactivity. This feedback loop enables dynamic correction of RTT measurements to account for caching engine delays and queuing effects
2Speed
If pipelined protocols with long pipelines are used, then data transmission speed improves, but RTT determination accuracy deteriorates
Solution Approach 1:
The patent performs preliminary actions by initializing the OC counter and FRTAIP timestamp before pipeline execution, and by pre-identifying the steady-state phase characteristics. This preparation enables accurate RTT measurement even in long pipelines by establishing reference points before the actual data transmission begins
Solution Approach 2:
The patent dynamically adapts the RTT measurement approach based on pipeline progression. The system transitions from tracking individual command delays in the initiation phase to aggregating pipeline-wide metrics in the steady-state phase, and finally to calculating completion delays in the termination phase. This dynamic adaptation maintains measurement accuracy regardless of pipeline length
3Reliability
If caching engine delays and network equipment inactivity are not accounted for, then the calculation process is simple, but the reliability of TS calculation deteriorates
Solution Approach 1:
The patent introduces intermediary variables (OC counter, IP flag, LRT timestamp) that act as mediators between the complex caching engine delays and the final RTT/TS calculations. These intermediaries capture the effects of caching engine processing and network equipment inactivity without requiring direct modification of the core calculation algorithms, thereby improving reliability while managing complexity
Solution Approach 2:
The patent changes key parameters during calculation based on observed conditions: switching between different RTT calculation formulas depending on whether the system is in pipeline initiation, steady-state, or completion phase. This parameter adaptation ensures that caching engine delays and network inactivity are properly accounted for in the final TS calculation
Data Source
AI summary
Methods and systems for determining latency in a communication network may be provided. A plurality of commands may be sent in the communication network. A response may be received for each of the plurality of commands from the communication network. A pipeline round trip time may be calculated for each of the plurality of commands based on a time of the response for each of the plurality of commands. A total round trip time for the plurality of commands may be calculated based on the calculated pipeline round trip time. A time saving may be calculated for at least one of the plurality of commands serviced locally.