{"id":2613,"date":"2022-11-03T15:38:38","date_gmt":"2022-11-03T15:38:38","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2022\/11\/03\/take-the-green-train-nvidia-bluefield-dpus-drive-data-center-efficiency\/"},"modified":"2022-11-03T15:38:38","modified_gmt":"2022-11-03T15:38:38","slug":"take-the-green-train-nvidia-bluefield-dpus-drive-data-center-efficiency","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2022\/11\/03\/take-the-green-train-nvidia-bluefield-dpus-drive-data-center-efficiency\/","title":{"rendered":"Take the Green Train: NVIDIA BlueField DPUs Drive Data Center Efficiency"},"content":{"rendered":"
\n

The numbers are in, and they paint a picture of data centers going a deeper shade of green, thanks to energy-efficient networks accelerated with data processing units (DPUs<\/a>).<\/p>\n

A suite of tests run with help from Ericsson, RedHat and VMware show power reductions up to 24% on servers using NVIDIA BlueField-2 DPUs<\/a>. In one case, they delivered 54x the performance of CPUs.<\/p>\n

The work, described in a recent whitepaper<\/a>, offloaded core networking jobs from power-hungry host processors to DPUs designed to run them more efficiently.<\/p>\n

Accelerated computing<\/a> with DPUs for networking, security and storage jobs is one of the next big steps for making data centers more power efficient. It\u2019s the latest of a handful of optimizations, described in the whitepaper, for data centers moving into the era of green computing<\/a>.<\/p>\n

DPUs Tested on VMware vSphere<\/b><\/h2>\n

Seeing the trend toward energy-efficient networks, VMware enabled DPUs to run its virtualization software, used by thousands of companies worldwide. NVIDIA has run several tests with VMware since its vSphere 8 software release<\/a> this fall.<\/p>\n

For example, on VMware vSphere Distributed Services Engine<\/a> \u2014\u00a0 software that offloads and accelerates networking and security functions using DPUs \u2014 BlueField-2 delivered higher performance while freeing up 20% of the CPU\u2019s resources required without DPUs.<\/p>\n

That means users can deploy fewer servers to run the same workload, or run more applications on the same servers.<\/p>\n

Power Costs Cut Nearly $2 Million<\/b><\/h2>\n

Few data centers face a more demanding job than those run by telecoms providers. Their networks shuttle every bit of data that smartphone users generate or request between their cellular networks and the internet.<\/p>\n

Researchers at Ericsson tested whether operators could reduce their power consumption on this massive workload using SmartNICs<\/a>, the network interface cards that handle DPU functions. Their test let CPUs slow down or sleep while an NVIDIA ConnectX SmartNIC<\/a> handled the networking tasks.<\/p>\n

The results, detailed in a recent article<\/a>, were stunning.<\/p>\n

Energy consumption of server CPUs fell 24%, from 190 to 145 watts on a fully loaded network. This single DPU application could cut power costs by nearly $2 million over three years for a large data center.<\/p>\n

\n

\"Ericsson

Ericsson ran the user-plane function for 5G networks on DPUs in three scenarios.<\/figcaption><\/figure>\n

In the article, Ericsson\u2019s CTO, Erik Ekudden, underscored the importance of the work.<\/p>\n

\u201cThere\u2019s a growing sense of urgency among communication service providers to find and implement innovative solutions that reduce network energy consumption,\u201d he wrote. And the DPU techniques \u201csave energy across a wide range of traffic conditions.\u201d<\/p>\n

70% Less Overhead, 54x More Performance<\/b><\/h2>\n

Results were even more dramatic for tests on Red Hat OpenShift<\/a>, used by half of all Fortune 500 banks, airlines and telcos to manage software containers.<\/p>\n

In the tests, BlueField-2 DPUs handled virtualization, encryption and networking jobs needed to manage these portable packages of applications and code.<\/p>\n

The DPUs slashed networking demands on CPUs by 70%, freeing them up to run other applications. What\u2019s more, they accelerated networking jobs by a whopping 54x.<\/p>\n

A technical blog<\/a> provides more detail on the tests.<\/p>\n

Speeding the Way to Zero Trust<\/b><\/h2>\n

Across every industry, businesses are embracing a philosophy of zero trust<\/a> to improve network security. So, NVIDIA tested IPsec, one of the most popular data center encryption protocols, on BlueField DPUs.<\/p>\n

The test showed data centers could improve performance and cut power consumption 21% for servers and 34% for clients on networks running IPsec on DPUs. For large data centers, that could translate to nearly $9 million in savings on electric bills over three years.<\/p>\n

NVIDIA and its partners continue to put DPUs to the test in an expanding portfolio of use cases, but the big picture is clear.<\/p>\n

\u201cIn a world facing rising energy costs and rising demand for green IT infrastructure, the use of DPUs will become increasingly popular,\u201d the whitepaper concludes.<\/p>\n

It\u2019s good to know the numbers, but seeing is believing. So apply<\/a> to run your own test of DPUs on VMware\u2019s vSphere.<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

https:\/\/blogs.nvidia.com\/blog\/2022\/11\/03\/bluefield-dpus-energy-efficiency\/<\/p>\n","protected":false},"author":0,"featured_media":2614,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[3],"tags":[],"_links":{"self":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/2613"}],"collection":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/comments?post=2613"}],"version-history":[{"count":0,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/2613\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media\/2614"}],"wp:attachment":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media?parent=2613"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/categories?post=2613"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/tags?post=2613"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}