DASH: a Recipe for a Flash-based Data Intensive Supercomputer

Abstract

Abstract-Data intensive computing can be defined as computation involving large datasets and complicated I/O patterns. Data intensive computing is challenging because there is a five-orders-of-magnitude latency gap between main memory DRAM and spinning hard disks; the result is that an inordinate amount of time in data intensive computing is spent accessing data on disk. To address this problem we designed and built a prototype data intensive supercomputer named DASH that exploits flash-based Solid State Drive (SSD) technology and also virtually aggregated DRAM to fill the "latency gap". DASH uses commodity parts including Intel® X25-E flash drives and distributed shared memory (DSM) software from ScaleMP®. The system is highly competitive with several commercial offerings by several metrics including achieved IOPS (input output operations per second), IOPS per dollar of system acquisition cost, IOPS per watt during operation, and IOPS per gigabyte (GB) of available storage. We present here an overview of the design of DASH, an analysis of its cost efficiency, then a detailed recipe for how we designed and tuned it for high data-performance, lastly show that running data-intensive scientific applications from graph theory, biology, and astronomy, we achieved as much as two orders-ofmagnitude speedup compared to the same applications run on traditional architectures.

Understanding the Impact of Emerging Non-Volatile Memories on High-Performance, IO-Intensive Computing

“…Our measurements and other recent work [13] shows that software RAID provides better performance for SSD-based arrays than hardware controllers, because the processors on hardware RAID controllers become a bottleneck. Therefore, we use software RAID for this array.…”

Parallel high-resolution climate data analysis using swift

“…• Dash [7], a cluster targeted for data-intensive computing at SDSC, consisting of 32 nodes each with dual quad-core 2.4 GHz Intel Nehalem processors and 48 GB of RAM per node. Half of Dash is configured as a traditional 16-node cluster but with a single 64 GB flash drive in each node.…”

Exploring the Future of Out-of-Core Computing with Compute-Local Non-Volatile Memory

“…Moving to the field, several research laboratories are making efforts to improve the performance of their supercomputers with the SSD-based storage. For instance, San Diego Supercomputing Center deployed SSDs in their supercomputer, Gordon [12,34,42], to reduce the latency gap between the memory and disks, as has the Tokyo Institute of Technology for their supercomputer, TSUBAME [27].…”