High Priority Applications Deserve Predictable High Performance

On April 2, 2019 Intel announced the next generation of foundational networking technology, the Intel® Ethernet 800 Series, with speeds up to 100GbE and many new innovative capabilities. One advanced capability that sets the new Intel Ethernet 800 series apart is Application Device Queues (ADQ). ADQ enables administrators to designate certain applications for priority network service in order to assure specific response times. It is like having an express lane reserved for carpools. ADQ provides fast and predictable data-transfer performance in a cost-effective manner that application developers and network operators can rely on and optimize their applications accordingly.

Traditionally, connectivity performance has been measured by latency and throughput – that is, the average response time and quantity of data movement. However, as data centers scale, predictability has become equally important, and is measured in terms of latency variability, or “jitter.” When more servers are added to support the growing network demands, jitter becomes a limiting factor to scalability and can result in a poor end-user experience. The goal of ADQ is to ensure that high-priority applications receive not only high performance, but predictable high performance through dramatically reduced jitter. At cloud scale, latency and high throughput must be coupled with predictability for companies to meet Service Level Agreements (SLAs).

Data Center Applications for ADQ

Figure 1. Initial Applications for ADQ in the Data Center

Initial deployments of ADQ are being targeted at the different tiers in the data center (Figure 1). Independent Software Vendors (ISVs) like Aerospike*, Redis Labs*, and NGINX* are excited about ADQ and Intel is working closely with these companies on optimizing ADQ for the Intel® Ethernet 800 Series.

Accelerated Packet Processing for Networking Applications

AF Express Data Path (AF_XDP) enables packet processing for cloud native solutions. Intel architects observed initial results of accelerating packet processing when using AF_XDP with ADQ on the Intel Ethernet 800 Series on a 2nd Generation Intel® Xeon® Scalable processor-based platform. Accelerating AF_XDP with ADQ lowers the number of CPU cores utilized by roughly half vs. without ADQ acceleration1, freeing up the processor cores used for packet processing for other more important tasks. The packet processing exhibits near-linear scaling as cores are added, making the aggregate performance across many cores highly scalable.

The Next Step in Network Predictability

ADQ improves application performance through higher response predictability, lower latency, and better throughout. Network predictability continues to grow in importance as data centers scale in the new data-centric era.  Initial applications that use ADQ have demonstrated great results, and there is great support from the ISV community. Most importantly, ADQ on the new Intel Ethernet 800 Series promises to be a significant step forward to serving end-users better and more consistently.

Read more about Intel Ethernet here: www.intel.com/Ethernet

1 Performance results are based on Intel internal testing as of January 2019, and may not reflect all publicly available security updates.  See configuration disclosure for details.  No product or component can be absolutely secure. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit www.intel.com/benchmarks.

Published on Categories Data Center NetworkingTags , , , ,
Patricia Kummrow

About Patricia Kummrow

Patty Kummrow is a VP in the Networking and Edge Group, and the GM of the Ethernet Products Group at Intel Corporation. She leads the strategy, architecture, development, manufacturing, and marketing of Intel® Ethernet Network Adapters, Controllers, and IPUs to enable next generation solutions used to accelerate networking, storage, and network security in data centers. Kummrow has two decades of experience in CPU design and technical leadership. She has led multiple teams developing Intel processors for data center, networking, storage, and autonomous driving applications. She holds a B.S. in Electrical Engineering from the University of Texas, and a M.S. in Technology Management from Walden University.