Aerospike First Commercial Database to Support Application Device Queues for More Predictable Application Response Times

For the next major milestone in the industry adoption of Application Devices Queues (ADQ), I’m excited to share news that Aerospike is the first commercial database to support the Intel® Ethernet 800 Series with ADQ technology in their 4.7 software release. First announced as a component of the Intel Ethernet 800 Series in April, ADQ is an open technology led by Intel for application-specific data packet queueing and steering across the network. ADQ enables administrators to designate certain applications for priority network service to enable more predictable application response times, toward the end result of providing greater consistency in meeting customer Service Level Agreements (SLAs).

As I explained earlier this year, the response time for an application is often limited by the slowest thread to respond, known in the industry as “tail latency.” The problem of tail latency worsens as systems scale. Though many elements make up the sequence that determines the response time, the network interface card and software stack can play a key role in impacting the total response time. The new Intel Ethernet 800 Series is designed to meet the most demanding server tasks with speeds up to 100GbE and many new innovative capabilities, such as ADQ.

Dedicated Lanes for Applications

Aerospike’s latest tests further prove how valuable ADQ will be for customers. Based on Aerospike’s testing in the Intel labs, ADQ can result in greater than 75% improved throughput and, most importantly, greater than 45% increased response time predictability for the Aerospike 4.7 next-generation NoSQL database running on a dual-socket 2nd Gen Intel® Xeon® Scalable processor-based server configured with two Intel Ethernet 800 Series network adapters with ADQ technology compared to the previously optimized approach of NUMA pinning.1 With Aerospike’s publicly available release, ADQ is now ready for organizations to begin their own proof-of-concept projects.

The Road to Hyperscale through both Storage and Network Improvements

“Traditionally, there are two bottlenecks that impede hyperscale data performance: storage and network,” said Srini Srinivasan, chief product officer and founder, Aerospike. “Aerospike was designed from the beginning for wire-line speed, and our new optimization for ADQ running on the Intel Ethernet 800 Series unlocks additional networking performance to power real-time enterprise applications at extreme scale in artificial intelligence, machine learning, and edge computing.”

Last year, Aerospike became the first commercially available open database to support Intel® Optane™ DC persistent memory on the 2nd Generation Intel Xeon Scalable processor-based platform. Now with support for ADQ, Aerospike is well positioned to help organizations overcome both the storage and network limitations to hyperscale performance.

Continued Collaboration

We’re fortunate to work with the Aerospike team to help customers get useful insights, real-time, out of big data.  We celebrate this key milestone of general availability of ADQ support in the Aerospike 4.7 release and look forward to our continued collaboration to support our customers with Intel hardware and Aerospike software for improved performance.

For more information about the benefits of ADQ, read Aerospike’s blog about the news or contact your Intel representative for how to start your own proof-of-concept project. For more on Intel’s vision for data center connectivity, visit intel.com/Ethernet.


Published on Categories Data Center NetworkingTags , , ,
Patricia Kummrow

About Patricia Kummrow

Patty Kummrow is a VP in the Data Center Group and the GM of the Ethernet Networking Division at Intel Corporation. She leads the design, development, manufacturing, and marketing of Intel® Ethernet Network Adapters, Controllers, Switch Silicon, along with 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, El Paso and a M.S. in Technology Management from Lehigh University.