Abstract | ||
---|---|---|
To increase datacenter energy efficiency, we need memory systems that keep pace with processor efficiency gains. Currently, servers use DDR3 memory, which is designed for high bandwidth but not for energy proportionality. A system using 20% of the peak DDR3 bandwidth consumes 2.3x the energy per bit compared to the energy consumed by a system with fully utilized memory bandwidth. Nevertheless, many datacenter applications stress memory capacity and latency but not memory bandwidth. In response, we architect server memory systems using mobile DRAM devices, trading peak bandwidth for lower energy consumption per bit and more efficient idle modes. We demonstrate 3-5x lower memory power, better proportionality, and negligible performance penalties for datacenter workloads.
|
Year | DOI | Venue |
---|---|---|
2012 | 10.1145/2366231.2337164 | ISCA |
Keywords | DocType | Volume |
memory system,memory capacity,memory bandwidth,mobile dram,lower memory power,ddr3 memory,towards energy-proportional datacenter memory,energy proportionality,datacenter energy efficiency,utilized memory bandwidth,high bandwidth,architect server memory system,energy efficient,memory management,mobile communication,data center,servers,bandwidth,energy conservation,stress | Conference | 40 |
Issue | ISSN | ISBN |
3 | 0163-5964 | 978-1-4503-1642-2 |
Citations | PageRank | References |
92 | 8.26 | 31 |
Authors | ||
6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Krishna T. Malladi | 1 | 249 | 18.37 |
Benjamin C. Lee | 2 | 931 | 51.85 |
Frank A. Nothaft | 3 | 92 | 8.26 |
Christos Kozyrakis | 4 | 5817 | 355.99 |
Karthika Periyathambi | 5 | 92 | 8.26 |
Mark Horowitz | 6 | 6374 | 1543.34 |