Flexible Hardware Acceleration for Instruction-Grain Program Monitoring

Chen, Guihai; Kozuch,; Strigkos,; Falsafi,; Gibbons,; Mowry,; Vignesh, Ramachandran; Ruwase,; Ryan, Martin; Vlachos,

doi:10.1109/isca.2008.20

Cited by 90 publications

(55 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This tight couplingwhere metadata is updated atomically with the corresponding application event-delays the application whenever a slow metadata operation arises, and requires either transactional memory [5] or special hardware support [6]. In contrast, asynchronous (a.k.a., "decoupled" or "loosely coupled") frameworks allow metadata updates (and evaluations for correctness) to lag behind the events that triggered them, 1 which typically results in much higher performance (because it is more tolerant of bursts in metadata computation) [7]. Table I compares monitoring frameworks where both the program and the lifeguard are parallel programs.…”

Section: A Lifeguard Frameworkmentioning

confidence: 99%

“…For brevity 7 In practice, our implementation sets thresholds for how long exploration of potential predecessors will continue, and explicitly tracks any early returns from exploration in a separate state called heuristic. 8 Valid ordering rules preclude an instruction repeating itself.…”

Section: B Resolving Transfer Functions To Metadatamentioning

confidence: 99%

“…Dynamic analysis tools [1,2,3,4,5,6,7,8,9,10] (aka "lifeguards") help programmers find bugs by performing sophisticated instruction-grain dynamic analysis of software as it executes. Lifeguards maintain metadata-shadow state about the application's memory and register state-to reason about whether the program is violating the lifeguard's model of correct execution.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Tracking and Reducing Uncertainty in Dataflow Analysis-Based Dynamic Parallel Monitoring

Goodstein

Gibbons

Kozuch

et al. 2015

2015 International Conference on Parallel Architecture and Compilation (PACT)

Self Cite

View full text Add to dashboard Cite

Dataflow analysis-based dynamic parallel monitoring (DADPM) is a recent approach for identifying bugs in parallel software as it executes, based on the key insight of explicitly modeling a sliding window of uncertainty across parallel threads. While this makes the approach practical and scalable, it also introduces the possibility of false positives in the analysis. In this paper, we improve upon the DADPM framework through two observations. First, by explicitly tracking new "uncertain" states in the metadata lattice, we can distinguish potential false positives from true positives. Second, as the analysis tool runs dynamically, it can use the existence (or absence) of observed uncertain states to adjust the tradeoff between precision and performance on-the-fly. For example, we demonstrate how the epoch size parameter can be adjusted dynamically in response to uncertainty in order to achieve better performance and precision than when the tool is statically configured. This paper shows how to adapt a canonical dataflow analysis problem (reaching definitions) and a popular security monitoring tool (TAINTCHECK) to our new uncertainty-tracking framework, and provides new provable guarantees that reported true errors are now precise.

show abstract

Section: A Lifeguard Frameworkmentioning

confidence: 99%

Section: B Resolving Transfer Functions To Metadatamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Tracking and Reducing Uncertainty in Dataflow Analysis-Based Dynamic Parallel Monitoring

Goodstein

Gibbons

Kozuch

et al. 2015

2015 International Conference on Parallel Architecture and Compilation (PACT)

Self Cite

View full text Add to dashboard Cite

show abstract

“…A dispatch-based approach called LogBased Architecture [9,10] and a distill-based approach [11,12,13] have been proposed. The main idea is to segregate the monitor and the application into different processes that run on different cores.…”

Section: Related Workmentioning

confidence: 99%

Adaptive Performance Monitoring for Embedded Multicore Systems

Shih

Lin

et al. 2011

2011 40th International Conference on Parallel Processing Workshops

View full text Add to dashboard Cite

With the advent of multicore processors, the performance of software has been elevated to new unforeseen heights via parallelization. However, this has not been achieved without new problems cropping up due to parallelization. One serious issue is the performance bottleneck due to cache misses or resource starvation, which is hard to detect in application software especially when the software has dynamically changing behavior. Performance monitors are usually employed for such purposes. Nevertheless, monitors have introduced their own computation and communication overheads, especially in embedded multicore systems. In this work, we try to estimate the effects of monitor overheads on different types of applications, such as CPU-bound and IO-bound tasks. Further, we give suggestions on the number and type of monitors to use for such embedded multicore applications. Besides trying to reduce monitor overheads, we also aim for the accuracy and the immediacy of the monitored information. Through a real-world example, namely digital video recording system, we demonstrate how different monitoring periods affect the tradeoff between accuracy and immediacy of the monitored information.

show abstract

“…For example, a software implementation for DIFT monitoring on a single core is reported to have an average slowdown of 3.6X even with aggressive optimizations [10]. The performance overheads of software monitoring can be mitigated using parallel processing [11], however, this entails using multiple processing cores to run each computation thread and is likely to increase power consumption by a factor of two or more.…”

Section: Introductionmentioning

confidence: 99%

High-performance parallel accelerator for flexible and efficient run-time monitoring

Deng

Suh

2012

IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 2012)

View full text Add to dashboard Cite

This paper proposes Harmoni, a high performance hardware accelerator architecture that can support a broad range of run-time monitoring and bookkeeping functions. Unlike custom hardware, which offers very little configurability after it has been fabricated, Harmoni is highly configurable and can allow a wide range of different hardware monitoring and bookkeeping functions to be dynamically added to a processing core even after the chip has already been fabricated. The Harmoni architecture achieves much higher efficiency than software implementations and previously proposed monitoring platforms by closely matching the common characteristics of run-time monitoring functions that are based on the notion of tagging. We implemented an RTL prototype of Harmoni and evaluated it with several example monitoring functions for security and programmability. The prototype demonstrates that the architecture can support a wide range of monitoring functions with different characteristics. Harmoni takes moderate silicon area, has very high throughput, and incurs low overheads on monitored programs.

show abstract

Flexible Hardware Acceleration for Instruction-Grain Program Monitoring

Cited by 90 publications

References 29 publications

Tracking and Reducing Uncertainty in Dataflow Analysis-Based Dynamic Parallel Monitoring

Tracking and Reducing Uncertainty in Dataflow Analysis-Based Dynamic Parallel Monitoring

Adaptive Performance Monitoring for Embedded Multicore Systems

High-performance parallel accelerator for flexible and efficient run-time monitoring

Contact Info

Product

Resources

About