A consistency architecture for hierarchical shared caches

The goal of this work is to design cache coherence protocols with many cores that can be verified with state-of-the-art automated verification methodologies. In particular, we focus on flat (non-hierarchical) coherence protocols, and we use a mostly-automated methodology based on parametric verification (PV). We propose several design guidelines that architects should follow if they want to design protocols that can be parametrically verified. We experimentally evaluate performance, storage overhead, and scalability of a protocol verified with PV compared to a highly optimized protocol that cannot be verified with PV.

show abstract

“…HCC [16] is organized hierarchically as a tree of caches. This tree organization facilitates verification of liveness and consistency.…”

Section: Related Workmentioning

confidence: 99%

PVCoherence: Designing flat coherence protocols for scalable verification

Zhang

Bingham

Erickson

et al. 2014

2014 IEEE 20th International Symposium on High Performance Computer Architecture (HPCA)

View full text Add to dashboard Cite

show abstract

“…Hierarchical or tree-based cache coherence organizations have been studied previously as a way to provide scalability in data storage and communication [Ladan-Mozes and Leiserson 2008]. The primary motivation in this work is to avoid timeouts and nacks by creating a unique path through the tree from each core to a given memory bank.…”

Section: Optimizing the Tag Directory Structurementioning

confidence: 99%

Using in-flight chains to build a scalable cache coherence protocol

Subramaniam¹,

Steely²,

Hasenplaugh

et al. 2013

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

As microprocessor designs integrate more cores, scalability of cache coherence protocols becomes a challenging problem. Most directory-based protocols avoid races by using blocking tag directories that can impact the performance of parallel applications. In this article, we first quantitatively demonstrate that state-of-the-art blocking protocols significantly constrain throughput at large core counts for several parallel applications. Nonblocking protocols address this throughput concern at the expense of scalability in the interconnection network or in the required resource overheads. To address this concern, we enhance nonblocking directory protocols by migrating the point of service of responses. Our approach uses in-flight chains of cores making parallel memory requests to incorporate scalability while maintaining high-throughput. The proposed cache coherence protocol called chained cache coherence, can outperform blocking protocols by up to 20% on scientific and 12% on commercial applications. It also has low resource overheads and simple address ordering requirements making it both a high-performance and scalable protocol. Furthermore, in-flight chains provide a scalable solution to building hierarchical and nonblocking tag directories as well as optimize communication latencies.

show abstract

“…Recently Ladan-Mozes and Leiserson [18] propose Hierarchical Cache Consistency (HCC), a deadlock-free, treebased coherence protocol that ensures forward progress in a fattree network. They do so by enumerating invariant properties that ensure all children in the tree are coherent with parents, forming the basis of their distributed coherence mechanism.…”

Section: Related Workmentioning

confidence: 99%

“…Despite having different goals, Fractal Coherence, HCC and MCP all share a similarity with respect to the permission inclusion property discussed by the authors of HCC in [18]. Parent/uppertier state in many respects can be viewed as a summary of the child/lower-tiers, having to have permissions at least as strong as the lower children/tiers'.…”

Section: Related Workmentioning

confidence: 99%

“…By expanding the scope of client agents to also monitor coherence realms in addition to processor caches, the coherence effort can be distributed over several protocols by layering the protocols in a tiered fashion. In order to enforce the permission-inclusion property described by Ladan-Mozes and Leiserson [18], the client agent must behave as a gateway for the manager of the coherence realm, restricting what permissions can be awarded, and taking action when permissions must be upgraded in the coherence realm before the manager can begin request resolution. The manager agents now must consult the gateway client before allocating permission, which in turn may recursively send another permission request to another manager- In both examples we have a top-tier coherence realm, A, that implements a low-overhead MEI protocol to manage two lower coherence realms, B and C, both implementing MOESI.…”

Section: Permission Hierarchy Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Manager-client pairing

Beu

Rosier

Conte

2011

Proceedings of the 44th Annual IEEE/ACM International Symposium on Microarchitecture

View full text Add to dashboard Cite

As technology continues to scale, the need for more sophisticated coherence management is becoming a necessity. The likely solution to this problem is the use of coherence hierarchies, analogous to how cache hierarchies have helped address the memory-wall problem in the past.Previous work in the construction of large-scale coherence protocols, however, demonstrates the complexity inherent to this design space.The difficulty with hierarchical coherence protocol design is the complexity increases exponentially with the increase in coherence states, due in turn to interactions between hierarchy tiers. Additionally, because of the large development investment, choices regarding coherence hierarchy are often made statically with little knowledge of how changes to the organization would affect the system. In this work, we present Manager-Client Pairing (MCP) as a unifying methodology for designing multi-tier coherence protocols by formally defining and limiting the interactions between levels within a coherence hierarchy to enable composition. Using MCP, we then implement a variety of hierarchical coherence protocol configurations for a 256-core system comprised of 4 64-core manycores, and provide insights into what impact different hierarchy depth and width choices can have on system performance.

show abstract

A consistency architecture for hierarchical shared caches

Cited by 14 publications

References 47 publications

PVCoherence: Designing flat coherence protocols for scalable verification

PVCoherence: Designing flat coherence protocols for scalable verification

Using in-flight chains to build a scalable cache coherence protocol

Manager-client pairing

Contact Info

Product

Resources

About