Wed 20 Jun 2018 11:25 - 11:50 at Grand Ballroom CD - Emerging Hardware Chair(s): Ryan R. Newton

Emerging Non-Volatile Memory (NVM) technologies offer high capacity and energy efficiency compared to DRAM, but suffer from limited write endurance and longer latencies. Prior work seeks the best of both technologies by combining DRAM and NVM in hybrid memories to attain low latency, high capacity, energy efficiency, and durability. Coarsegrained hardware and OS optimizations then spread writes out (wear-leveling) and place highly mutated pages in DRAM to extend NVM lifetimes. Unfortunately even with these coarsegrained methods, popular Java applications exact impractical NVM lifetimes of 4 years or less. This paper shows how to make hybrid memories practical, without changing the programming model, by enhancing garbage collection in managed language runtimes. We find object write behaviors offer two opportunities: (1) 70% of writes occur to newly allocated objects, and (2) 2% of objects capture 81% of writes to mature objects. We introduce writerationing garbage collectors that exploit these fine-grained behaviors. They extend NVM lifetimes by placing highly mutated objects in DRAM and read-mostly objects in NVM. We implement two such systems. (1) Kingsguard-nursery places new allocation in DRAM and survivors in NVM, reducing NVM writes by 5× versus NVM only with wear-leveling. (2) Kingsguard-writers (KG-W) places nursery objects in DRAM and survivors in a DRAM observer space. It monitors all mature object writes and moves unwritten mature objects from DRAM to NVM. Because most mature objects are unwritten, KG-W exploits NVM capacity while increasing NVM lifetimes by 11×. It reduces the energy-delay product by 32% over DRAM-only and 29% over NVM-only. This work opens up new avenues for making hybrid memories practical.

Wed 20 Jun

Displayed time zone: Eastern Time (US & Canada) change

11:00 - 12:15
Emerging HardwarePLDI Research Papers at Grand Ballroom CD
Chair(s): Ryan R. Newton Indiana University
11:00
25m
Talk
Persistency for Synchronization-Free Regions
PLDI Research Papers
Vaibhav Gogte University of Michigan, USA, Stephan Diestelhorst ARM Research, UK, William Wang Arm Research, UK, Satish Narayanasamy University of Michigan, Peter M. Chen University of Michigan, USA, Thomas F. Wenisch University of Michigan, USA
Media Attached
11:25
25m
Talk
Write-Rationing Garbage Collection for Hybrid Memories
PLDI Research Papers
Shoaib Akram Ghent University, Jennifer B. Sartor Vrije Universiteit Brussel, Kathryn S McKinley Google, Lieven Eeckhout Ghent University, Belgium
Media Attached
11:50
25m
Talk
Mapping Spiking Neural Networks onto a Manycore Neuromorphic Architecture
PLDI Research Papers
Chit-Kwan Lin Intel Labs, n.n., Andreas Wild Intel Labs, n.n., Tsung-Han Lin Intel Labs, n.n., Gautham N. Chinya Intel Labs, n.n., Mike Davies Intel Labs, n.n., Hong Wang Intel Labs, n.n.
Media Attached