Sandvine (TSX:SVC) has announced that the new Intel Xeon Scalable processor platform will provide their Policy Traffic Switch Virtual Series with up to 60 percent more packet processing power.
As quoted in the press release:
As a member of the Intel® Network Builders ecosystem, Sandvine had early access to Intel Xeon Gold 6150 processor based platforms. After extensive in-house testing using Sandvine’s current PTS Virtual Series software build, Sandvine measured packet inspection throughput improvements of up to 60% over a comparable reference platform using Intel® Xeon® E5-2699 v3 processors. The most significant improvements came when inspecting small (64- and 128-byte) packets, which are typically the biggest challenge from a performance perspective. The improvements mean that communications service providers (CSPs) will now need to utilize fewer vCPUs in their virtual workload environment to achieve a given level of throughput and to enable Sandvine’s network policy control platform at line rate speeds.
Sandvine’s Virtual Series products are currently deployed in the control and data planes of CSPs of all sizes, including TextNow, an MVNO, Bakcell, a tier-1 mobile operator, and Digicel, a multi-national operator group. When these and other Sandvine’s Virtual Series customers migrate their deployments to systems that take advantage of the Intel Xeon Scalable processor capabilities, the improved packet processing performance will allow them to implement significantly more complex network policy control use cases without the need for additional software licenses.
“Without any optimization to our PTS Virtual Series software, the Intel Xeon Scalable processor will provide our customers with the ability to do significantly more with the Sandvine platform without the need for additional vCPUs,” said Don Bowman, CTO. “Later this year when we update our Virtual Series products to take advantage of the Intel® AVX-512 instruction set featured in the new processor, we will also be able to enhance our cyber security solutions with innovative new features based on deep machine learning and neural-network acceleration.”