(opens in new tab)<\/span><\/a>Derek Chiou, a Bing hardware architect, discusses the benefits of the collaboration.<\/p>\n\u201cThe partnership between Doug and his team at Microsoft Research and Bing has been fantastic and has resulted in significant results that will have real impact on Bing,\u201d Chiou says. \u201cThe factor of two throughput improvement demonstrated in the pilot means we can do the same amount of work with half the number of servers or double the amount of work with the same number of servers\u2014or some mix of the two.<\/p>\n
\u201cThose kinds of numbers are especially significant at the scale of a datacenter. The potential benefits go beyond simple dollars. To give some examples, Bing\u2019s ranking could be further enhanced to provide an even better customer experience, power could be saved, and the size of the datacenters could be reduced. The strength of the pilot results have led to Bing deploying this technology in one datacenter for customers, starting in early 2015.\u201d<\/p>\n
As the ISCA paper notes, FPGAs have become powerful computing devices in recent years, making them particularly suited for use as fine-grained accelerators.<\/p>\n
\u201cWe designed a platform that permits the software in the cloud, which is inherently programmable, to partner with programmable hardware,\u201d Burger says. \u201cYou can move functions into custom hardware, but rather than burning them into fixed chips [application-specific integrated circuits], we map them to Altera FPGAs, which can run hardware designs but can be changed by reconfiguring the FPGA.<\/p>\n
\u201cWe\u2019ve demonstrated a \u2018programmable hardware\u2019 enhanced cloud, running smoothly and reliably at large scale.\u201d<\/p>\n
In the evaluation deployment outlined in the paper, the reconfigurable fabric\u2014interconnected nodes linked by high-bandwidth connections\u2014was tested on a collection of 1,632 servers to measure its efficacy in accelerating the workload of a production web-search service. The results were impressive: a 95 percent improvement in throughput at a latency comparable to a software-only solution. With an increase in power consumption and total per-server cost increase of less than 30 percent, the net results deliver substantial savings and efficiencies.<\/p>\n
The results demonstrated the project\u2019s capability to run stably for long periods, and all the stages in the pipeline exceeded the overall throughput goal. In addition, a service to handle failures quickly reconfigures the fabric after errors or machine failures.<\/p>\n
The ISCA paper concludes by underscoring the belief that distributed reconfigurable fabrics will play a critical role as server performance increases level off. Such techniques could become indispensable to datacenter managers balancing their conflicting goals.<\/p>\n
\u201cThis portends a future where systems are specialized dynamically by compiling a good chunk of demanding workloads into hardware,\u201d Burger says. \u201cI would imagine that a decade hence, it will be common to compile applications into a mix of programmable hardware and programmable software.<\/p>\n
\u201cThis is a radical shift that will offer continued performance improvements past the end of Moore\u2019s Law as we move more and more of our applications and services into hardware.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"
Posted by Rob Knies Operating a datacenter at web scale requires managing many conflicting requirements. The ability to deliver computation at a high level and speed is a given, but because of the demands such a facility must meet, a datacenter also needs flexibility. Additionally, it must be efficient in its use of power, keeping […]<\/p>\n","protected":false},"author":30766,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"footnotes":""},"categories":[194470,194463],"tags":[200189,186604,194924,201063,201231,195319,195383,195583,186412,202067,202113,196808,203507,203753],"research-area":[13547],"msr-region":[],"msr-event-type":[],"msr-locale":[268875],"msr-post-option":[],"msr-impact-theme":[],"msr-promo-type":[],"msr-podcast-series":[],"class_list":["post-509","post","type-post","status-publish","format-standard","hentry","category-computer-architecture","category-systems","tag-a-reconfigurable-fabric-for-accelerating-large-scale-datacenter-services","tag-bing","tag-catapult","tag-computer-systems-and-networking","tag-datacenter","tag-derek-chiou","tag-doug-burger","tag-field-programmable-gate-array","tag-fpga","tag-international-symposium-on-computer-architecture","tag-isca","tag-peter-lee","tag-reconfigurable-fabric","tag-server","msr-research-area-systems-and-networking","msr-locale-en_us"],"msr_event_details":{"start":"","end":"","location":""},"podcast_url":"","podcast_episode":"","msr_research_lab":[],"msr_impact_theme":[],"related-publications":[],"related-downloads":[],"related-videos":[],"related-academic-programs":[],"related-groups":[],"related-projects":[171431],"related-events":[],"related-researchers":[],"msr_type":"Post","byline":"","formattedDate":"June 16, 2014","formattedExcerpt":"Posted by Rob Knies Operating a datacenter at web scale requires managing many conflicting requirements. The ability to deliver computation at a high level and speed is a given, but because of the demands such a facility must meet, a datacenter also needs flexibility. Additionally,…","locale":{"slug":"en_us","name":"English","native":"","english":"English"},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/509"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/users\/30766"}],"replies":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/comments?post=509"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/509\/revisions"}],"predecessor-version":[{"id":235607,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/509\/revisions\/235607"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=509"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/categories?post=509"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/tags?post=509"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=509"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=509"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=509"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=509"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=509"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=509"},{"taxonomy":"msr-promo-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-promo-type?post=509"},{"taxonomy":"msr-podcast-series","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-podcast-series?post=509"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}