Minerva: Accelerating Data Analysis in Next-Generation SSDs
Authors:
- Arup DeADArup De
- Maya GokhaleMGMaya Gokhale
- Rajesh GuptaRGRajesh Gupta
- Steven SwansonSSSteven Swanson
- 1 April 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 9-16
Abstract
Emerging non-volatile memory (NVM) technologies have DRAM-like latency with storage-like density, offering unique capability to analyze large data sets significantly faster than flash or disk storage. However, the hybrid nature of these NVM technologies such as phase-change memory (PCM) make it difficult to use them to best advantage in the memory-storage hierarchy. These NVMs lack the fast write latency required of DRAM and are thus not suitable as DRAM equivalent on the memory bus, yet their low latency even in random access patterns is not easily exploited over an I/O bus. In this work, we describe an FPGA-based system to execute application-specific operations in the NVM controller and evaluate its performance on two micro benchmarks and a key valuestore. Our system Minerva extends the conventional solid-state drive (SSD) architecture to offload data or I/O intensive application code to the SSD to exploit the low latency and high internal bandwidth of NVMs. Performing computation in the FPGA-based NVM storage controller significantly reduces data traffic between the host and storage and serves as an offload engine for data analysis workloads. A runtime library enables the programmer to offload computations to the SSD without dealing with the complications of the underlying architecture and inter-controller communication management. We have implemented a prototype of Minerva on the BEE3FPGA system. We compare the performance of Minerva to a state of the art PCIe-attached PCM-based SSD. Minerva improves performance by an order of magnitude on two micro benchmarks. Minerva based key-value store performs up to 5.2 M get operations/s and 4.0 M set operations/s which is7.45脳 and 9.85脳 higher than the PCM-based SSD that uses the conventional I/O architecture. This huge improvement comes from the reduction of data transfer between the storage to the host and the FPGA-based data processing in the SSD.Keywords
This publication has 12 references indexed in Scilit:
- Providing safe, user space access to fast, solid state disks2012
- Benchmarking cloud serving systems with YCSB2010
- Enhancing lifetime and security of PCM-based main memory with start-gap wear leveling2009
- PDRAM2009
- Scalable high performance main memory system using phase-change memory technology2009
- Architecting phase change memory as a scalable dram alternative2009
- Dynamo2007
- Active disks1998
- A case for intelligent disks (IDISKs)ACM SIGMOD Record, 1998
- Performance Analysis of Alternative Database Machine ArchitecturesIEEE Transactions on Software Engineering, 1982