Time sensitive queue priority scheduling method and device based on data plane

Abstract

The invention relates to a time-sensitive queue priority scheduling method and device based on a data plane, and the method comprises the steps: receiving a data package, analyzing the data package, and obtaining network state information; calculating to obtain urgency parameters according to the network state information; presetting a priority threshold, and performing grade division on the data packet according to the urgency parameter to obtain a scheduling priority; and scheduling the data packet according to the scheduling priority. According to the invention, the real-time performance and the overall efficiency of network transmission are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of network resource allocation, and in particular relates to a method and device for scheduling time-sensitive queue priority based on a data plane. Background Art

[0002] The traditional Internet provides best-effort services, and in terms of QoS assurance, it also mostly focuses on resources in spatial dimensions such as paths and bandwidth, including queue scheduling, resource allocation and other mechanisms, which are also optimized at the spatial level. With the rise of time-sensitive applications such as industrial Internet, remote control, and real-time interaction, optimization of the time dimension has become a new breakthrough point for network upgrades. Low latency in the network is particularly important for many emerging services and applications such as drones, industrial automation, and self-driving cars. In response to the above needs, international standards organizations are developing new technologies to meet the requirements of these deterministic applications. Deterministic network technology is used to provide real-time data transmission and ensure deterministic communication service quality such as ultra-low upper bounds of latency, jitter, and packet loss rate, controllable upper and lower bounds of bandwidth, and ultra-high lower bounds of reliability; time-sensitive network technology achieves zero congestion packet loss transmission through a series of protocol standards, provides low latency and jitter with upper bounds, and provides deterministic services for latency-sensitive traffic.

[0003] There are many studies devoted to reducing data center network latency, most of which are centered around the goal of minimizing flow completion time. For example, D-SRTF uses sent data to estimate the remaining flow size, while taking into account the network's available bandwidth to optimize the minimum flow completion time. However, under the premise that the flow size is unknown, using the statistical information of the sent data to estimate the remaining flow completion time is an approximate estimate based on statistics, which has certain problems. At the same time, the flow-based queue priority scheduling algorithm will determine different priorities for data packets in the same flow according to the different transmission stages of the flow, which will cause the problem of increased delay jitter. In addition, traditional priority scheduling algorithms are deployed at the terminal, and the congestion parameters fed back by the network are used to estimate the current state of the network, which cannot accurately perceive the network status. Summary of the invention

[0004] In view of the above deficiencies in the prior art, an object of the present invention is to provide a method and device for time-sensitive queue priority scheduling based on a data plane.

[0005] The present invention provides a time-sensitive queue priority scheduling method based on a data plane, comprising:

[0006] S1: receiving a data packet, parsing the data packet, and obtaining network status information;

[0007] S2: Calculate and obtain an urgency parameter according to the network status information;

[0008] S3: Preset a priority threshold, classify the data packets according to the urgency parameter, and obtain a scheduling priority;

[0009] S4: Schedule the data packets according to the scheduling priority.

[0010] According to a data plane-based time-sensitive queue priority scheduling method provided by the present invention, before step S1, the method further includes:

[0011] S111: inserting network status information into a header field of a data packet passing through a preceding switch through a programmable data plane device.

[0012] According to a time-sensitive queue priority scheduling method based on a data plane provided by the present invention, the network status information includes:

[0013] Hop delay information and total hop count information.

[0014] According to a time-sensitive queue priority scheduling method based on a data plane provided by the present invention, step S111 specifically includes:

[0015] Based on the header field of the data packet, when bit0 is 1, the current switch ID is embedded in the data packet;

[0016] When bit2 is 1, the hopping delay information is embedded in the data packet.

[0017] According to a time-sensitive queue priority scheduling method based on a data plane provided by the present invention, step S1 further comprises:

[0018] S121: Capture data packets through the switch network interface and perform integrity check on the data packets;

[0019] S122: Locate the data packet that has passed the integrity check to obtain the field to be parsed;

[0020] S123: parse and extract the to-be-parsed fields to obtain network status information.

[0021] According to a time-sensitive queue priority scheduling method based on a data plane provided by the present invention, the expression of the urgency parameter in step S2 is:

[0022]

[0023] in, is the switch index value, is the total number of switches, The first time a data packet arrives at the network The urgency parameter when jumping the switch, The upper limit of the transmission time of a data packet in the network. It is the sum of the transmission delays of the data packet at each switch in the network.

[0024] According to a time-sensitive queue priority scheduling method based on a data plane provided by the present invention, in step S3, the expression of the priority threshold is:

[0025]

[0026] in, is the index value corresponding to the set priority threshold. is the highest initial priority, is a constant parameter.

[0027] A second aspect of the present invention provides a time-sensitive queue priority scheduling device based on a data plane, comprising:

[0028] Parsing module: used for receiving data packets, parsing the data packets, and obtaining network status information;

[0029] Calculation module: used for calculating and obtaining the urgency parameter according to the network status information;

[0030] A classification module: used to classify data packets according to the urgency parameter based on the priority threshold to obtain a scheduling priority;

[0031] Scheduling module: used to schedule data packets according to the scheduling priority.

[0032] A third aspect of the present invention provides a time-sensitive queue priority scheduling device based on a data plane, comprising:

[0033] A memory and at least one processor, wherein instructions are stored in the memory;

[0034] At least one of the processors calls the instructions in the memory to enable a data plane-based time-sensitive queue priority scheduling device to execute a data plane-based time-sensitive queue priority scheduling method as described in any one of the above items.

[0035] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed by a processor, a time-sensitive queue priority scheduling method based on a data plane as described in any one of the above items is implemented.

[0036] The present invention provides a method, device, equipment and storage medium for scheduling a time-sensitive queue priority based on a data plane. Through the method for scheduling a time-sensitive queue priority based on a data plane, the urgency parameter of a data packet can be calculated in real time according to network status information, and the data packet can be prioritized and scheduled according to the parameter, ensuring that high-priority data packets can be processed and transmitted faster, thereby improving the real-time and overall efficiency of network transmission; secondly, the present invention uses a programmable data plane device to embed network status information into the header field of the data packet, so that the data packet can carry more useful information during transmission, which not only enhances the adaptability of the network to dynamic changes, but also improves the flexibility of network scheduling; thirdly, the present invention can easily adapt to different network environments and application requirements by adjusting the calculation method of the priority threshold and the urgency parameter; in addition, the present invention can more effectively utilize network resources by accurately calculating the urgency parameter of the data packet and prioritizing it, avoiding low-priority data packets from occupying too many resources and causing network congestion, which helps to improve the throughput and stability of the network and ensure the timely transmission of key data; in the process of receiving and parsing the data packet, the present invention also ensures the integrity and accuracy of the data packet through integrity checking. This inspection mechanism helps prevent data packets from being tampered with or damaged during transmission, thereby improving the security and reliability of network transmission. The time-sensitive queue priority scheduling method based on the data plane of the present invention is easy to implement and expand. Through simple configuration and adjustment, it can be applied to different types of network devices and scenarios to meet various complex network requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are only used to illustrate specific embodiments and are not considered to limit the present invention. In the entire drawings, the same reference symbols represent the same components. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0038] Figure 1 A schematic diagram of a flow chart of a time-sensitive queue priority scheduling method based on a data plane provided by an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of a time-sensitive queue priority scheduling device based on a data plane provided by an embodiment of the present invention.

[0040] Reference numerals:

[0041] 100, parsing module; 200, calculation module; 300, partitioning module; 400, scheduling module. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work should fall within the scope of protection of the present invention.

[0043] Furthermore, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts disclosed in the present invention.

[0044] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the orientation or position relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside", etc. is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. The terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0045] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of methods and systems consistent with some aspects of the present invention as detailed in the appended claims.

[0046] In order to better understand the embodiments of the present invention, the research background of the present invention is first explained below.

[0047] In data center networks, many studies focus on minimizing flow completion time. However, current research on improving data center network latency performance has the following problems:

[0048] There is an inaccurate problem in estimating the remaining time of a flow using the sent data. Whether estimating the remaining flow size by the number of sent data packets or estimating the urgency of the flow by the flow sent time, under the premise that the flow size is unknown, it is an inaccurate approximate estimation based on statistical significance. For example, the D-SRTF algorithm may misjudge a long flow as a short flow in the early stage of flow transmission.

[0049] Delay jitter caused by flow-based scheduling algorithms. The D-SRTF and Aemon algorithms are both flow-based queue priority scheduling algorithms. Whether estimating the remaining flow size based on the sent data or introducing the urgency parameter based on the flow transmission time, each flow is used as the basic scheduling unit and the priority of each flow is determined at the terminal. Flow-based scheduling algorithms (D-SRTF and AEMON) determine different priorities for data packets at different flow transmission stages (Priority 0 has the highest priority), which will cause the problem of increased delay jitter. Therefore, the ideal scheduling algorithm should have better granularity to solve the problem of increased delay jitter caused by different priorities at different flow transmission stages.

[0050] The problem of not being able to perceive changes in network conditions in real time. The priority of the above algorithms is determined at the terminal, and its priority is only related to the stage of this stream transmission, and has nothing to do with the network conditions during the transmission process. Among them, D-SRTF uses the parameters that reflect network congestion in DCTCP at the terminal to estimate the current network available bandwidth. This parameter that reflects network congestion will have a certain lag after end-to-end transmission, and cannot accurately represent the current network congestion situation. Therefore, the traditional scheduling algorithm has limited network perception capabilities and cannot achieve accurate real-time control according to changes in network conditions. The scheduling decision of the ideal scheduling algorithm should not be completed only at the terminal, but should be distributed and deployed in the network, so that its decision-making layer is closer to the network side, and a network-assisted scheduling algorithm is realized, so that the algorithm has a more efficient network status perception capability.

[0051] The embodiments of the present invention are described below with reference to the accompanying drawings.

[0052] like Figure 1 As shown, the present invention provides a time-sensitive queue priority scheduling method based on a data plane, comprising:

[0053] S1: Receive a data packet, parse the data packet, and obtain network status information.

[0054] Furthermore, in step S1, in-band network status information is first obtained. Specifically, when the data packet passes through the current switch, the programmable data plane INT architecture is required to insert the hop delay information and the total number of hops of the passed switches into the normal data packet header field in an in-band manner. When the data packet passes through the subsequent switch, the data packet header field can be parsed to obtain the network status information.

[0055] Wherein, before step S1, the method further includes:

[0056] S111: inserting network status information into a header field of a data packet passing through a preceding switch through a programmable data plane device.

[0057] The network status information includes:

[0058] Hop delay information and total hop count information.

[0059] Furthermore, in the header field of the data packet, the TotalHopCnt field counts the hop count information of the switches in the network through which the data packet passes, and the Instruction Bitmap field identifies the acquired network status information, so the network status information to be inserted can be embedded later according to the field information.

[0060] Wherein, step S111 specifically includes:

[0061] Based on the header field of the data packet, when the bit0 position is 1, the current switch ID is embedded in the data packet; when the bit2 position is 1, the hop delay information is embedded in the data packet.

[0062] During the insertion process, when bit0 is set to 1, the ID of the switch it passes through will be embedded in the data packet. When bit2 is set to 1, the data plane device will embed the hop delay information into the packet header field. Therefore, when the data packet reaches the next hop switch, the data packet contains the hop delay information of each switch it has experienced in the network.

[0063] Wherein, step S1 further comprises:

[0064] S121: Capture data packets through the switch network interface and perform integrity check on the data packets.

[0065] In step S121, when a data packet enters the switch, the network interface of the switch first captures and receives the data packet. Before the data packet is passed to the custom parser, some preprocessing steps may be required, such as data packet integrity check, deduplication, etc. After preprocessing, the data packet is passed to the custom parser.

[0066] S122: Locate the data packet that has passed the integrity check and obtain the field to be parsed.

[0067] After receiving the data packet, the custom parser is first initialized to prepare the resources and data structures required for parsing, and then locates the field storing the hop delay information and the total number of switch hops in the data packet through the above-mentioned bit positions.

[0068] S123: parse and extract the to-be-parsed fields to obtain network status information.

[0069] After locating the relevant fields, the parser will extract the information in these fields. For the hop delay information, it is the difference between one or more timestamps. For the total number of switch hops information, it is a counter value. After extracting the information, the parser will perform verification to ensure the accuracy and completeness of the information, including checking the format and range of the data.

[0070] S2: Calculate and obtain an urgency parameter according to the network status information.

[0071] In step S2, the urgency of each data packet is calculated based on the network status information (hop delay, total number of hops, and number of hops experienced) obtained in step S1. The specific urgency parameter calculation method is as follows.

[0072] The expression of the urgency parameter in step S2 is:

[0073]

[0074] in, is the switch index value, is the total number of switches, The first time a data packet arrives at the network The urgency parameter when jumping the switch, The upper limit of the transmission time of a data packet in the network. is the sum of the transmission delays of the data packet on each switch in the network. The number of switch hops that the data packet has passed through in the network.

[0075] S3: Preset a priority threshold, classify the data packets according to the urgency parameter, and obtain a scheduling priority.

[0076] Wherein, in step S3, the expression of the priority threshold is:

[0077]

[0078] in, is the index value corresponding to the set priority threshold. is the highest initial priority, is a constant parameter.

[0079] In this embodiment, a total of 8 priorities are designed, corresponding to 8 queues respectively, so the threshold calculation formula in the above is i ∈[0,6] , and are all preset fixed values. In this embodiment, , .

[0080] According to the urgency parameter of each data packet calculated in step S2, a priority is assigned to each data packet. is set to priority 0 when the urgency parameter is greater than Less than , and so on, where priority 0 represents the highest priority, i.e. packets assigned to this priority will be forwarded first, and priority 7 represents the lowest priority, i.e. packets assigned to this priority will be forwarded last.

[0081] S4: Schedule the data packets according to the scheduling priority.

[0082] The scheduling in step S4 is based on the priority assigned to each data packet in step S3, that is, the data packets are placed in the corresponding queues according to their priorities. Since the priorities have been divided according to the urgency parameters, the scheduler can ensure that high-priority data packets, that is, data packets with higher urgency, are processed or forwarded first.

[0083] like Figure 2 As shown, the present invention also provides a time-sensitive queue priority scheduling device based on a data plane, comprising:

[0084] Parsing module 100: used to receive data packets, parse the data packets, and obtain network status information;

[0085] Calculation module 200: used to calculate and obtain an urgency parameter according to the network status information;

[0086] The classification module 300 is used to classify the data packets according to the urgency parameter based on the priority threshold to obtain the scheduling priority;

[0087] Scheduling module 400: configured to schedule data packets according to the scheduling priority.

[0088] The present invention also provides a time-sensitive queue priority scheduling device based on a data plane, comprising:

[0089] A memory and at least one processor, wherein instructions are stored in the memory;

[0090] At least one of the processors calls the instructions in the memory to enable a data plane-based time-sensitive queue priority scheduling device to execute a data plane-based time-sensitive queue priority scheduling method as described in any one of the above items.

[0091] The present invention also provides a computer-readable storage medium, on which instructions are stored. When the instructions are executed by a processor, a time-sensitive queue priority scheduling method based on a data plane as described in any one of the above items is implemented.

[0092] Furthermore, a time-sensitive queue priority scheduling device based on a data plane provided by the present invention may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU), for example, one or more processors and memories, one or more storage media for storing applications or data, such as one or more mass storage devices, wherein the memories and storage media may be short-term storage or permanent storage, and the program stored in the storage medium may include one or more modules, each module may include a series of instruction operations in a time-sensitive queue priority scheduling device based on a data plane, and further, the processor may be configured to communicate with the storage medium to execute a series of instruction operations in the storage medium on a time-sensitive queue priority scheduling device based on a data plane.

[0093] It may also include one or more power supplies, one or more wired or wireless network interfaces, one or more input and output interfaces, and one or more operating systems, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that the data plane-based time-sensitive queue priority scheduling device structure provided by the present invention does not constitute a limitation on the device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently.

[0094] In a specific embodiment, after simulation experiments, the present invention simulates and implements a typical three-layer network adopted by a data center network, and uses the ECMP protocol for network load balancing. The experiments show that compared with existing solutions, the present invention has achieved significant performance improvements in flow completion time, delay jitter and deadline completion rate. Among them, compared with the PISA algorithm, the flow completion time of the present invention is reduced by an average of 5%-8.1%, and in terms of delay jitter, the average jitter of the present invention is reduced by 12.6%-18%, and in terms of deadline completion rate, the present invention is improved by an average of 23%-43%.

[0095] The present invention provides a method, device, equipment and storage medium for scheduling time-sensitive queues based on the data plane, which solves the problem that traditional network service quality assurance algorithms are generally deployed at network terminals. The purpose of traditional methods is to reduce the complexity of network equipment (including switches and routers) as much as possible, thereby reducing the cost of network deployment. In this context, since the real network status cannot be obtained in real time, the algorithm often uses various methods to estimate the network status, but cannot schedule data packets in real time during the transmission process. In order to cope with the problem of increasingly complex network capability requirements for network-borne services in recent years, increasing the computing power of the network itself has become a natural solution, and intelligent networks have become an inevitable choice.

[0096] Programmable data plane technology is a protocol-independent network forwarding technology that opens up network programming interfaces and greatly promotes the space for innovation in network technology. With the rapid development of programmable data plane technology and smart network card technology, network devices have opened up programmable capabilities and have certain computing capabilities. Therefore, it also provides new ideas and methods for solving traditional network topics such as network service quality assurance. Distributed deployment of related algorithms in network element devices can increase the perception and feedback capabilities of network status. Obviously, the "network-assisted" scheduling algorithm is a feasible idea and solution, so the present invention proposes a distributed queue priority scheduling method implemented in the data plane, which realizes a time-sensitive data plane technology.

[0097] The present invention provides a time-sensitive queue priority scheduling method, device, equipment and storage medium based on the data plane, which realizes a fine-grained queue priority scheduling algorithm at the packet level of the programmable data plane, greatly reduces the delay jitter between data packets, and at the same time utilizes the in-band computing capability of the programmable data plane to determine the priority of the current data packet according to the network status information during the data packet transmission process. Compared with the traditional algorithm method of estimating the network status at the terminal according to the congestion parameters fed back by the network, the perception ability of the network status is improved.

[0098] The present invention implements a distributed queue priority scheduling algorithm with more accurate network status perception capability. The algorithm of the present invention is not only deployed on the terminal, but also makes full use of the computing power of the programmable data plane switch itself to deploy the queue priority scheduling algorithm on the network side in a distributed manner. This solves the problem that traditional algorithms cannot accurately and real-time perceive the network status.

[0099] The present invention also implements a packet-level fine-grained time-sensitive scheduling algorithm. By using the in-band computing capability of the programmable data plane, under the constraint of the upper limit of delay, the delay urgency of each data packet on the remaining transmission nodes is calculated according to the delay consumed by the data packet transmission, and queue priority scheduling is performed based on this. Compared with the traditional algorithm, fine-grained control at the packet level is achieved, and on the basis of ensuring the completion of transmission under the constraint of the upper limit of delay, network delay jitter can be reduced at the same time.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the above embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or replacements that can be easily thought of by those skilled in the art within the technical scope disclosed in the present invention should be covered within the protection scope of the present invention.

Claims

1. A time-sensitive queue priority scheduling method based on a data plane, characterized in that: include: S1: receiving a data packet, parsing the data packet, and obtaining network status information; S2: Calculate and obtain an urgency parameter according to the network status information; S3: Preset a priority threshold, classify the data packets according to the urgency parameter, and obtain a scheduling priority; S4: Schedule the data packets according to the scheduling priority.

2. A time-sensitive queue priority scheduling method based on data plane according to claim 1, characterized in that: Before step S1, the method further includes: S111: inserting network status information into a header field of a data packet passing through a preceding switch through a programmable data plane device.

3. A time-sensitive queue priority scheduling method based on data plane according to claim 2, characterized in that: The network status information includes: Hop delay information and total hop count information.

4. A time-sensitive queue priority scheduling method based on data plane according to claim 3, characterized in that: Step S111 specifically includes: Based on the header field of the data packet, when bit0 is 1, the current switch ID is embedded in the data packet; When bit2 is 1, the hopping delay information is embedded in the data packet.

5. The method for scheduling priority of a time-sensitive queue based on a data plane according to claim 1, characterized in that: Step S1 further comprises: S121: Capture data packets through the switch network interface and perform integrity check on the data packets; S122: Locate the data packet that has passed the integrity check to obtain the field to be parsed; S123: parse and extract the to-be-parsed fields to obtain network status information.

6. A method for scheduling time-sensitive queue priority based on data plane according to claim 1, characterized in that: The expression of the urgency parameter in step S2 is: ； in, is the switch index value, is the total number of switches, The first time a data packet arrives at the network The urgency parameter when jumping the switch, The upper limit of the transmission time of a data packet in the network. It is the sum of the transmission delays of the data packet at each switch in the network.

7. The method for scheduling priority of a time-sensitive queue based on a data plane according to claim 1, characterized in that: In step S3, the expression of the priority threshold is: ； in, is the index value corresponding to the set priority threshold. is the highest initial priority, is a constant parameter.

8. A time-sensitive queue priority scheduling device based on a data plane, characterized in that: include: Parsing module: used for receiving data packets, parsing the data packets, and obtaining network status information; Calculation module: used for calculating and obtaining the urgency parameter according to the network status information; A classification module: used to classify data packets according to the urgency parameter based on the priority threshold to obtain a scheduling priority; Scheduling module: used to schedule data packets according to the scheduling priority.

9. A time-sensitive queue priority scheduling device based on a data plane, characterized in that: include: A memory and at least one processor, wherein instructions are stored in the memory; At least one of the processors calls the instructions in the memory to enable a data plane-based time-sensitive queue priority scheduling device to execute a data plane-based time-sensitive queue priority scheduling method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed by the processor, a time-sensitive queue priority scheduling method based on a data plane is implemented as described in any one of claims 1-7.

Images