XChange: A market-based approach to scalable dynamic multi-resource allocation in multicore architectures

Abstract

Efficiently allocating shared on-chip resources across cores is critical to optimize execution in chip multiprocessors (CMPs). Techniques proposed in the literature often rely on global, centralized mechanisms that seek to maximize system throughput. Global optimization may hurt scalability: as more cores are integrated on a die, the search space grows exponentially, making it harder to achieve optimal or even acceptable operating points at run-time without incurring significant overheads.In this paper, we propose XChange, a novel CMP resource allocation mechanism that delivers scalable high throughput and fairness. Through XChange, the CMP functions as a market, where each shared resource is assigned a price which changes over time, and each core seeks to maximize its own utility, by bidding for these shared resources. Because each core works largely independently, the resource allocation becomes a scalable, mostly distributed decision-making process. In addition, by distributing the resources proportionally to the bids, the system avoids unfairness, treating each core in an unbiased manner.Our evaluation shows that, using detailed simulations of a 64-core CMP configuration running a variety of multiprogrammed workloads, the proposed XChange mechanism improves system throughput (weighted speedup) by about 21% on average, and fairness (harmonic speedup) by about 24% on average, compared with equal-share on-chip cache and power distribution. On both metrics, that is at least about twice as much improvement over equal-share as a state-ofthe-art centralized allocation scheme. Furthermore, our results show that XChange is significantly more scalable than the state-of-the-art centralized allocation scheme we compare against.

SEDEA: A Sensible Approach to Account DRAM Energy in Multicore Systems

“…However, the cache and memory bandwidth requirements of a task are complementary [26]. This claim builds on the observation that, when a task has allocated a sufficiently large cache partition, it needs less memory bandwidth.…”

Research Problems and Opportunities in Memory Systems

“…Resources are then allocated appropriately to different applications so that a global system performance metric is optimized. Another example of such a scheme is a recent work by Wang and Martinez [191] that employs a market-dynamics-inspired mechanism to coordinate allocation decisions across resources. Approaches to coordinate resource allocation and scheduling decisions across multiple resources in the memory system, whether they use machine learning, game theory, or feedback based control, are a promising research topic that offers ample scope for future exploration.…”

Last level cache partitioning via multiverse thread classification

“…Wang and Martinez allocate resources that include LLC space among multiple cores with a market-based approach, where each core tries to obtain maximum utility with a given budget [13]. Resource prices are determined by total demand on the resource, and cores iteratively update their bids according to the changes in prices until the system converges to a balance.…”