AstriFlash A Flash-Based System for Online Services
Authors:
- 1 February 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 81-93
Abstract
Modern datacenters host datasets in DRAM to offer large-scale online services with tight tail-latency requirements. Unfortunately, as DRAM is expensive and increasingly difficult to scale, datacenter operators are forced to consider denser storage technologies. While modern flash-based storage exhibits μs-scale access latency, which is well within the tail-latency constraints of many online services, traditional demand paging abstraction used to manage memory and storage incurs high overheads and prohibits flash usage in online services. We introduce AstriFlash, a hardware-software co-design that tightly integrates flash and DRAM with ns-scale overheads. Our evaluation of server workloads with cycle-accurate full-system simulation shows that AstriFlash achieves 95% of a DRAM-only system’s throughput while maintaining the required 99th-percentile tail latency and reducing the memory cost by 20x.Keywords
Funding Information
- Qualcomm
- National Research Foundation of Korea
This publication has 51 references indexed in Scilit:
- Knights Landing: Second-Generation Intel Xeon Phi ProductIEEE Micro, 2016
- Unified address translation for memory-mapped SSDs with FlashMap2015
- Scale-out NUMA2014
- Die-stacked DRAM caches for servers2013
- The tail at scaleCommunications of the ACM, 2013
- Fundamental Latency Trade-off in Architecting DRAM Caches: Outperforming Impractical SRAM-Tags with a Simple and Practical Design2012
- Towards energy-proportional datacenter memory with mobile DRAM2012
- Clearing the clouds2012
- The context-switch overhead inflicted by hardware interrupts (and the enigma of do-nothing loops)2007
- Hints for computer system design1983