Improving Multiprocessor Performance with Coarse-Grain Coherence Tracking

Cantin, Jason F.; Lipasti, Mikko H.; Smith, James E.

doi:10.1109/isca.2005.31

Cited by 70 publications

(29 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As used in CGCT [3], this technique can reduce snoops effectively on clean data. However, it may increase the latencies of L2 misses for content-shared data, since it does not check whether there are cached copies, which can be obtained by fast on-chip cache-to-cache transfers.…”

Section: Support For Content-based Sharingmentioning

confidence: 99%

“…RegionScount maintains region-based coherence filters at requesting nodes to avoid broadcasting snoop requests for private data [2]. Coarse-grain coherence tracking (CGCT) also uses additional coarse-grained coherence tags for each cache and tracks the private or shared states of regions, in addition to the conventional cacheline-unit coherence [3]. Snoop requests are either broadcast or sent directly to the memory depending on the coarse grain states.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Virtual Snooping Coherence for Multi-Core Virtualized Systems

Kim

Park

Kim³

et al. 2016

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Proliferation of virtualized systems opens a new opportunity to improve the scalability of multi-core architectures. Among the scalability bottlenecks in multi-cores, cache coherence has been one of the most critical problems. Although snoop-based protocols have been dominating commercial multi-core designs, it has been difficult to scale them for more cores, as snooping protocols require high network bandwidth and power consumption for snooping all the caches. In this paper, we propose a novel snoop-based cache coherence protocol, called virtual snooping, for virtualized multi-core architectures. Virtual snooping exploits memory isolation across virtual machines and prevents unnecessary snoop requests from crossing the virtual machine boundaries. Each virtual machine becomes a virtual snoop domain, consisting of a subset of the cores in a system. Although the majority of virtual machine memory is isolated, sharing of cachelines across VMs still occur. To address such data sharing, this paper investigates three factors, data sharing through the hypervisor, virtual machine relocation, and content-based sharing. In this paper, we explore the design space of virtual snooping with experiments on emulated and real virtualized systems including the mechanisms and overheads of the hypervisor. In addition, the paper discusses the scheduling impact on the effectiveness of virtual snooping.

show abstract

Section: Support For Content-based Sharingmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Virtual Snooping Coherence for Multi-Core Virtualized Systems

Kim

Park

Kim³

et al. 2016

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

show abstract

“…Coarse-grain coherence tracking [7] and RegionScout [33] both propose mechanisms to reduce coherence traffic in broadcast-based systems by managing coherence at a coarser granularity than a line. While these techniques can reduce storage costs, both mechanisms impose restrictions on alignment and sizing of coherence regions and may lead to increased message traffic; both are situations we wish to avoid with Cohesion.…”

Section: Hardware Schemesmentioning

confidence: 99%

Cohesion

Kelm

Johnson

Tuohy

et al. 2010

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

Two broad classes of memory models are available today: models with hardware cache coherence, used in conventional chip multiprocessors, and models that rely upon software to manage coherence, found in compute accelerators. In some systems, both types of models are supported using disjoint address spaces and/or physical memories. In this paper we present Cohesion, a hybrid memory model that enables fine-grained temporal reassignment of data between hardware-managed and software-managed coherence domains, allowing a system to support both. Cohesion can be used to dynamically adapt to the sharing needs of both applications and runtimes. Cohesion requires neither copy operations nor multiple address spaces.Cohesion offers the benefits of reduced message traffic and on-die directory overhead when software-managed coherence can be used and the advantages of hardware coherence for cases in which software-managed coherence is impractical. We demonstrate our protocol using a hierarchical, cached 1024-core processor with a single address space that supports both software-enforced coherence and a directory-based hardware coherence protocol. Relative to an optimistic, hardware-coherent baseline, a realizable Cohesion design achieves competitive performance with a 2× reduction in message traffic, 2.1× reduction in directory utilization, and greater robustness to on-die directory capacity.

show abstract

“…A preliminary evaluation of RegionScout appears in [26]. Cantin, Lipast and Smith have also proposed exploiting coarse sharing for snoop coherence bandwidth reduction [8].…”

Section: (A) Locall2mentioning

confidence: 99%

RegionScout

Moshovos¹

2005

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

It has been shown that many requests miss in all remote nodes in shared memory multiprocessors. We are motivated by the observation that this behavior extends to much coarser grain areas of memory. We define a region to be a continuous, aligned memory area whose size is a power of two and observe that many requests find that no other node caches a block in the same region even for regions as large as 16K bytes. We propose RegionScout, a family of simple filter mechanisms that dynamically detect most non-shared regions. A node with a RegionScout filter can determine in advance that a request will miss in all remote nodes. RegionScout filters are implemented as a layered extension over existing snoop-based coherence systems. They require no changes to existing coherence protocols or caches and impose no constraints on what can be cached simultaneously. Their operation is completely transparent to software and the operating system. RegionScout filters require little additional storage and a single additional global signal. These characteristics are made possible by utilizing imprecise information about the regions cached in each node. Since they rely on dynamically collected information RegionScout filters can adapt to changing sharing patterns. We present two applications of RegionScout: In the first RegionScout is used to avoid broadcasts for non-shared regions thus reducing bandwidth. In the second RegionScout is used to avoid snoop induced tag lookups thus reducing energy.

show abstract

Improving Multiprocessor Performance with Coarse-Grain Coherence Tracking

Abstract: (and up to 21.7%).

Cited by 70 publications

References 22 publications

Virtual Snooping Coherence for Multi-Core Virtualized Systems

Virtual Snooping Coherence for Multi-Core Virtualized Systems

Cohesion

RegionScout

Contact Info

Product

Resources

About