Jason Mars
University of Virginia, Computer Science, Graduate Student
Abstract Hybrid processors are HW/SW co-designed processors that leverage blocked-execution, the execution of regions of instructions as atomic blocks, to facilitate aggressive speculative optimization. As we move to a multicore hybrid... more
Abstract Hybrid processors are HW/SW co-designed processors that leverage blocked-execution, the execution of regions of instructions as atomic blocks, to facilitate aggressive speculative optimization. As we move to a multicore hybrid design, fine grained conflicts for shared data can violate the atomicity requirement of these blocks and lead to expensive squashes and rollbacks.
Bubble-Up's key insight is that predicting the performance interference of co-running applications can be decoupled into measuring the pressure an application generates on the memory subsystem and measuring how much an application suffers... more
Bubble-Up's key insight is that predicting the performance interference of co-running applications can be decoupled into measuring the pressure an application generates on the memory subsystem and measuring how much an application suffers from different levels of pressure. The underlying hypothesis is that both the pressure and sensitivity can be quantified using a common pressure metric. Having such a metric reduces the complexity of colocation analysis.
Abstract Due to the complexity and the massive scale of modern warehouse scale computers (WSCs), it is challenging to quantify the performance impact of individual microarchitectural properties and the potential optimization benefits in... more
Abstract Due to the complexity and the massive scale of modern warehouse scale computers (WSCs), it is challenging to quantify the performance impact of individual microarchitectural properties and the potential optimization benefits in the production environment. As a result of these challenges, there is currently a lack of understanding of the microarchitecture-workload interaction, leaving potentially significant performance on the table.
Abstract With the shift to chip multiprocessors, managing shared resources has become a critical issue in realizing their full potential. Previous research has shown that thread mapping is a powerful tool for resource management. However,... more
Abstract With the shift to chip multiprocessors, managing shared resources has become a critical issue in realizing their full potential. Previous research has shown that thread mapping is a powerful tool for resource management. However, the difficulty of simultaneously managing multiple hardware resources and the varying nature of the workloads have impeded the efficiency of thread mapping algorithms.
Abstract As multicore processors with expanding core counts continue to dominate the server market, the overall utilization of the class of datacenters known as warehouse scale computers (WSCs) depends heavily on colocation of multiple... more
Abstract As multicore processors with expanding core counts continue to dominate the server market, the overall utilization of the class of datacenters known as warehouse scale computers (WSCs) depends heavily on colocation of multiple workloads on each server to take advantage of the computational power provided by modern processors. However, many of the applications running in WSCs, such as websearch, are user-facing and have quality of service (QoS) requirements.
Abstract Precisely predicting performance degradation due to colocating multiple executing applications on a single machine is critical for improving utilization in modern warehouse-scale computers (WSCs). Bubble-Up is the first mechanism... more
Abstract Precisely predicting performance degradation due to colocating multiple executing applications on a single machine is critical for improving utilization in modern warehouse-scale computers (WSCs). Bubble-Up is the first mechanism for such precise prediction. As opposed to over-provisioning machines, Bubble-Up enables the safe colocation of multiple workloads on a single machine for Web service applications that have quality of service constraints, thus greatly improving machine utilization in modern WSCs.
Page 1. 1 Rethinking the Architecture of Warehouse-Scale Computers A Dissertation Presented to the faculty of the School of Engineering and Applied Science University of Virginia In Partial Fulfillment of the requirements for the Degree... more
Page 1. 1 Rethinking the Architecture of Warehouse-Scale Computers A Dissertation Presented to the faculty of the School of Engineering and Applied Science University of Virginia In Partial Fulfillment of the requirements for the Degree Doctor of Philosophy Computer Science by Jason Mars May 2012 Page 2.
Abstract The overhead of test coverage analysis is dominated by monitoring the application, which is traditionally performed using instrumentation. However, instrumentation can prohibitively increase the time and especially the memory... more
Abstract The overhead of test coverage analysis is dominated by monitoring the application, which is traditionally performed using instrumentation. However, instrumentation can prohibitively increase the time and especially the memory overhead of an application. As an alternative to instrumentation, we explore how recent hardware advances can be leveraged to improve the overheads of test coverage analysis. These hardware advances include hardware performance monitors and multicore technology.
Abstract As much of the world's computing continues to move into the cloud, the overprovisioning of computing resources to ensure the performance isolation of latency-sensitive tasks, such as web search, in modern datacenters is a major... more
Abstract As much of the world's computing continues to move into the cloud, the overprovisioning of computing resources to ensure the performance isolation of latency-sensitive tasks, such as web search, in modern datacenters is a major contributor to low machine utilization.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Abstract Achieving effective online adaptation for natively executed applications has proved quite challenging and to date has not been widely adopted. Traditionally, to enable online adaptation for native binary applications, a run-time... more
Abstract Achieving effective online adaptation for natively executed applications has proved quite challenging and to date has not been widely adopted. Traditionally, to enable online adaptation for native binary applications, a run-time layer is added that virtualizes the execution of the application by performing dynamic binary to binary translation. This virtual layer injects trampolines and instrumentation into the translated code to maintain control of the application. This approach adds significant overhead and complexity to the application ...