{"id":572415,"date":"2019-03-13T09:58:54","date_gmt":"2019-03-13T16:58:54","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?p=572415"},"modified":"2019-03-13T09:58:54","modified_gmt":"2019-03-13T16:58:54","slug":"researchers-seek-to-simplify-the-complex-in-cloud-computing","status":"publish","type":"post","link":"https:\/\/www.microsoft.com\/en-us\/research\/blog\/researchers-seek-to-simplify-the-complex-in-cloud-computing\/","title":{"rendered":"Researchers seek to simplify the complex in cloud computing"},"content":{"rendered":"
<\/p>\n
From February 26\u201328, researchers gathered in Boston for the 16th USENIX Symposium on Networked Systems Design and Implementation (NSDI) (opens in new tab)<\/span><\/a>, one of the top conferences in the networking and systems field. Microsoft, a silver sponsor of the event, was represented by researchers serving on the program committee, as well as those presenting papers (opens in new tab)<\/span><\/a>, including two research teams using novel abstractions to empower and better serve cloud users.<\/p>\n \u201cBoth papers describe new ways to cope with the ever-increasing scale and complexity of what it means to do state-of-the-art computing in the cloud,\u201d said Thomas Moscibroda, Microsoft Partner Research Scientist, Azure Compute (opens in new tab)<\/span><\/a>.<\/p>\n With their respective work, the teams seek to simplify the underlying operations\u2014or what Microsoft Principal Scientist Konstantinos Karanasos (opens in new tab)<\/span><\/a>, co-author on the other paper, calls \u201cthe magic\u201d\u2014to deliver a more efficient and seamless user experience.<\/p>\n Field programmable gate arrays (FPGAs) are becoming widely used in today\u2019s data centers (opens in new tab)<\/span><\/a>. These reprogrammable circuits combine the advantages of hardware speed while offering some of the flexibility that makes software ideal for programming. But taking advantage of their full potential at cloud computing scale has been extremely challenging for several reasons, and researchers in the Networking Research Group at Microsoft Research Asia (opens in new tab)<\/span><\/a>, in collaboration with engineering leaders in Microsoft Azure (opens in new tab)<\/span><\/a>, are hoping to change that by addressing one such obstacle: the absence of an efficient, reliable, easy-to-use communications layer.<\/p>\n In their paper \u201cDirect Universal Access: Making Data Center Resources Available to FPGA,\u201d (opens in new tab)<\/span><\/a> they present a new communications architecture, one that Microsoft Researcher Peng Cheng (opens in new tab)<\/span><\/a> and his co-authors liken to the Internet Protocol or the operating system of a computer.<\/p>\n \u201cOur challenge has been, how do we provide a software-like IP layer inside this hardware-based platform,\u201d said Cheng, adding that the goal is a unified platform.<\/p>\n Currently, communication between pairs of FPGAs and other data center resources, such as CPUs, GPUs, memory, and storage, is complex, making programming large-scale heterogenous applications impractical and, at times, nearly impossible.<\/p>\n There are several reasons for this, the researchers explain in their paper: First, the communications paradigms used for connecting resources that are local to a server and resources that are remote\u2014that is, located on a different server in the data center\u2014are different and use vastly different communications stacks. Secondly, resources are named in a way that is specific to the server they live on. And lastly, current FPGA architecture is inefficient when it comes to multiplexing multiple diverse communications links to different local and remote resources.<\/p>\n <\/p>\nDirect Universal Access: A communications architecture<\/h3>\n