OS-controlled cache predictability for real-time systems

Liedtke, Jochen; Härtig, Hermann; Hohmuth, Michael

doi:10.1109/rttas.1997.601360

Cited by 124 publications

(116 citation statements)

References 12 publications

Supporting

Mentioning

116

Contrasting

Order By: Relevance

“…We reserved cache partitions 0-7 (CBM bitmask 0×000FF) to CPU1 and partitions 8-15 (CBM bitmask 0×0FF00) to CPU2. We flushed the entire cache initially, and mitigated potential interference to CPU1 and CPU2 by moving all system services to the remaining cores and assigning to them the remaining partitions (partitions [16][17][18][19]. We created a periodic task that sequentially accesses a 4MB array.…”

Section: Cache Lookup Controlmentioning

confidence: 99%

See 1 more Smart Citation

vCAT: Dynamic Cache Management Using CAT Virtualization

Thi²,

Phan

et al. 2017

2017 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS)

View full text Add to dashboard Cite

This paper presents vCAT, a novel design for dynamic shared cache management on multicore virtualization platforms based on Intel's Cache Allocation Technology (CAT). Our design achieves strong isolation at both task and VM levels through cache partition virtualization, which works in a similar way as memory virtualization, but has challenges that are unique to cache and CAT. To demonstrate the feasibility and benefits of our design, we provide a prototype implementation of vCAT, and we present an extensive set of microbenchmarks and performance evaluation results on the PARSEC benchmarks and synthetic workloads, for both static and dynamic allocations. The evaluation results show that (i) vCAT can be implemented with minimal overhead, (ii) it can be used to mitigate shared cache interference, which could have caused task WCET increased by up to 7.2 x, (iii) static management in vCAT can increase system utilization by up to 7 x compared to a system without cache management; and (iv) dynamic management substantially outperforms static management in terms of schedulable utilization (increase by up to 3 x in our multi-mode example use case). Abstract-This paper presents vCAT, a novel design for dynamic shared cache management on multicore virtualization platforms based on Intel's Cache Allocation Technology (CAT). Our design achieves strong isolation at both task and VM levels through cache partition virtualization, which works in a similar way as memory virtualization, but has challenges that are unique to cache and CAT. To demonstrate the feasibility and benefits of our design, we provide a prototype implementation of vCAT, and we present an extensive set of microbenchmarks and performance evaluation results on the PARSEC benchmarks and synthetic workloads, for both static and dynamic allocations. The evaluation results show that (i) vCAT can be implemented with minimal overhead, (ii) it can be used to mitigate shared cache interference, which could have caused task WCET increased by up to 7.2×, (iii) static management in vCAT can increase system utilization by up to 7× compared to a system without cache management; and (iv) dynamic management substantially outperforms static management in terms of schedulable utilization (increase by up to 3× in our multi-mode example use case). Disciplines Computer Engineering | Computer Sciences

show abstract

Section: Cache Lookup Controlmentioning

confidence: 99%

“…Existing work in this domain relies on software-based solutions, such as page coloring [17], for cache partitioning. Most previous solutions focus on improving the average performance of the system (e.g., [10,11,24]) and thus are not suitable for real-time systems that require worst-case performance guarantees.…”

Section: Introductionmentioning

confidence: 99%

vCAT: Dynamic Cache Management Using CAT Virtualization

Thi²,

Phan

et al. 2017

2017 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS)

View full text Add to dashboard Cite

show abstract

“…Kaseridis et al [11] propose bandwidth-aware memory sub-system management for avoiding resource contention. Various approaches are also proposed to manage LLC [5,8,14,15,20,31,32,33,34]. In particular, Qureshi et al [32] design a utility-based cache partitioning scheme that allocates appropriate cache resources based on application miss rate monitored through dedicated hardware.…”

Section: Related Workmentioning

confidence: 99%

Going vertical in memory management: Handling multiplicity by multi-policy

Liu

Cui

et al. 2014

2014 ACM/IEEE 41st International Symposium on Computer Architecture (ISCA)

View full text Add to dashboard Cite

show abstract

“…These works deals only with fixed priority schedulers. Other works try to eliminate or reduce the inter-task interference by using hardware and software techniques [7][8][9][10], but they do not face the intrinsic interference problem. Additionally, in some cases, the extrinsic interference is only reduced, and therefore, the predictability problem of the cache-related preemption delay remains unresolved.…”

Section: Introductionmentioning

confidence: 99%

Schedulability Analysis in EDF Scheduler with Cache Memories

Campoy

Sáez

Perles

et al. 2004

Real-Time and Embedded Computing Systems and Applications

View full text Add to dashboard Cite

Abstract. Cache memories can improve computer performance, but its unpredictable behaviour makes difficult to use them in hard real-time systems. Classical analysis techniques are not sufficient to accomplish schedulability analysis, and new hardware resources or complex analysis algorithms are needed. This work presents a comprehensive method to obtain predictability on the use of caches in real-time systems using an EDF scheduler. Reaching a predictable cache, schedulability analysis can be accomplished in a simple way through conventional algorithms. At the moment, this is the first approach to consider cache in this kind of scheduler. The method is based in the use of locking caches and genetic algorithms. Locking caches allows to load and lock cache contents, ensuring its remains unchanged. Genetic algorithms help to select the cache content that offers the best performance. Experimental results indicate that this scheme is fully predictable, and this predictability is reached with no performance loss for around 60% of cases.

show abstract

OS-controlled cache predictability for real-time systems

Abstract: Abstract

Cited by 124 publications

References 12 publications

vCAT: Dynamic Cache Management Using CAT Virtualization

vCAT: Dynamic Cache Management Using CAT Virtualization

Going vertical in memory management: Handling multiplicity by multi-policy

Schedulability Analysis in EDF Scheduler with Cache Memories

Contact Info

Product

Resources

About