This paper presents an overview of a storage system software stack designed specifically for extreme scale data centric computing. The storage system software stack is designed to work for next generation storage syst...
详细信息
ISBN:
(纸本)9781538610442
This paper presents an overview of a storage system software stack designed specifically for extreme scale data centric computing. The storage system software stack is designed to work for next generation storage system architectures that will need to incorporate multiple data storage device technologies. We also envision such datacentric storage systems to have in-storage compute capability to make it suitable for dealing with data analytics, and, pre/post processing steps in datacentric workflows. The storage system software stack indeed needs to accommodate such capability.
In the era of exploding internet usage, social and mobile, enterprises are facing both the challenges and business opportunities that are introduced by Big data, which has the characteristics of high volume, high velo...
详细信息
ISBN:
(纸本)9781450316361
In the era of exploding internet usage, social and mobile, enterprises are facing both the challenges and business opportunities that are introduced by Big data, which has the characteristics of high volume, high velocity, and high variety. Big data and the emergence of Internet-facing workloads will blur the separation between traditional transactional and analytics workloads. To extract business value and make actionable insight from the unprecedented volume of the data with the agility required from the business, it requires transformational innovations from many fronts. For example, in data management layer, how unstructured data is stored and retrieved efficiently, how data-intensive analytic computation can be done on commercial systems effectively, how the distributed cache should be designed to make use of the latest network protocols so the network-connected memory data can be accessed remotely and seamlessly. Moreover, the trend also motivates many architectural and technological advancement, such as moving from a transaction-centric to a data-centric architecture that supports extreme low and predicable latency, massive scale-out, high concurrency, and real-time situational awareness and analytics, and that requires orders of magnitude improvement over existing systems across each of these characteristics. At the same time, new applications in the Mobile and social space leverage new open source software stacks written in multiple programming languages, e.g., Java, JavaScript, Ruby, PHP, where the developer chooses the best tool for the job. How a polyglot runtime platform can be built that serves as a best practice platform for the programmers' community and in the meantime, optimized for enterprises with elastic, lightweight, resilient, agile runtime for business computing. Last, but not least, how the benchmarks should be enriched to measure the new runtimes, new data-centric systems and *** this talk, I will talk about some of the researc
The conventional approach of moving stored data to the CPU for computation has become a major performance bottleneck for emerging scale-out data-intensive applications due to their limited data reuse. At the same time...
详细信息
ISBN:
(纸本)9781538673768
The conventional approach of moving stored data to the CPU for computation has become a major performance bottleneck for emerging scale-out data-intensive applications due to their limited data reuse. At the same time, the advancement in integration technologies have made the decade-old concept of coupling compute units close to the memory (called Near-Memory computing) more viable. Processing right at the home of data can completely diminish the data movement problem of data-intensive applications. This paper focuses on analyzing and organizing the extensive body of literature on near-memory computing across various dimensions: starting from the memory level where this paradigm is applied, to the granularity of the application that could be executed on the near-memory units. We highlight the challenges as well as the critical need of evaluation methodologies that can be employed in designing these special architectures. Using a case study, we present our methodology and also identify topics for future research to unlock the full potential of near-memory computing.
暂无评论