Abstract

Big data applications like graph processing are highly imposed on memory capacity. Byte-addressable non-volatile memory (NVM) technologies can offer much larger memory capacity, lower cost per bit relative to traditional DRAM. They are expected to play a crucial role in mitigating I/O operations for big data processing. However, since the NVMs show higher access latency and lower bandwidth compared with DRAM, it is still challenging to fully exploit the advantages of both the DRAM and NVM for graph processing. In this paper, we propose NGraph, a new parallel graph processing framework specially designed for hybrid memory systems. According to different access patterns of graph data, NGraph exploits memory heterogeneity-aware data placement strategies to avoid random accesses and frequent updates to NVM. NGraph partitions graph by destination vertices and exploits a task decomposition scheme to avoid data contention between multicores. Meanwhile, the NGraph balances the execution time of parallel graph data processing on multicores through a work-stealing strategy. Moreover, the NGraph also proposes software-based data pre-fetching to improve cache hit rate, and supports huge page to reduce address translation overhead. We evaluate NGraph using a hybrid memory emulator. The experimental results show that NGraph can achieve up to 48.28% performance improvement for several typical benchmarks compared with the state-of-the-art systems Ligra and Polymer.

Highlights

  • Social network, road network and biological network contain a lot of valuable information, and mining this information from massive data plays an important role in enterprises’ decision making

  • Non-Volatile Memory (NVM) shows lower bandwidth and higher access latency than that of DRAM, we find that a hybrid memory system has a potential to improve the performance of graph processing by up to 6 times compared to an out-of-core

  • This paper aims to design a graph processing system on such hybrid memory system consisting of a small size of DRAM (GB) and a relatively large amount of NVM (TB)

Read more

Summary

Introduction

Road network and biological network contain a lot of valuable information, and mining this information from massive data plays an important role in enterprises’ decision making. Because traditional DRAM technologies generally feature low memory density, high cost and power consumption, they cannot satisfy the increasing requirement of main memory for high-performance large graph processing. Emerging Non-Volatile Memory (NVM) [9] technologies such as PCM [10] and ReRAM [11] have the potential to radically change the landscape of memory systems They generally show higher memory density, lower energy consumption and lower cost per GB than DRAM. NVM TECHNOLOGIES Non-Volatile Memories(NVM) such as MRAM [18], F-RAM [19], STT-MRAM [20], PCM [10], [21], [22] and RRAM [11] are characterized with byte-addressable, nearzero static energy consumption and high memory density. The NVM represents Intel’s newly released Optane DC Persistent Memory DIMMs [24]

Objectives
Methods
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call