arXiv subjects

Tan N. Le

Publications and source records attributed to Tan N. Le.

3 recordsLinked to original sources

arXiv2020

Flex: Closing the Gaps between Usage and Allocation

Data centers are giant factories of Internet data and services. Worldwide data centers consume energy and emit emissions more than airline industry. Unfortunately, most of data centers are significantly underutilized. One of the major reasons is the big gaps between the real usage and the provisioned resources because users tend to over-estimate their demand and data center operators often rely on users' requests for resource allocation. In this paper, we first conduct an in-depth analysis of a Google cluster trace to unveil the root causes for low utilization and highlight the great potential to improve it. We then developed an online resource manager Flex to maximize the cluster utilization while satisfying the Quality of Service (QoS). Large-scale evaluations based on real-world traces show that Flex admits up to 1.74x more requests and 1.6x higher utilization compared to tradition schedulers while maintaining the QoS.

cs.DC↗

arXiv2019

BoPF: Mitigating the Burstiness-Fairness Tradeoff in Multi-Resource Clusters

Simultaneously supporting latency- and throughout-sensitive workloads in a shared environment is an increasingly more common challenge in big data clusters. Despite many advances, existing cluster schedulers force the same performance goal - fairness in most cases - on all jobs. Latency-sensitive jobs suffer, while throughput-sensitive ones thrive. Using prioritization does the opposite: it opens up a path for latency-sensitive jobs to dominate. In this paper, we tackle the challenges in supporting both short-term performance and long-term fairness simultaneously with high resource utilization by proposing Bounded Priority Fairness (BoPF). BoPF provides short-term resource guarantees to latency-sensitive jobs and maintains long-term fairness for throughput-sensitive jobs. BoPF is the first scheduler that can provide long-term fairness, burst guarantee, and Pareto efficiency in a strategyproof manner for multi-resource scheduling. Deployments and large-scale simulations show that BoPF closely approximates the performance of Strict Priority as well as the fairness characteristics of DRF. In deployments, BoPF speeds up latency-sensitive jobs by 5.38 times compared to DRF, while still maintaining long-term fairness. In the meantime, BoPF improves the average completion times of throughput-sensitive jobs by up to 3.05 times compared to Strict Priority.

cs.DC↗

arXiv2019

Geographically Coordinated Primary Frequency Control Technical Report

Primary Frequency Control (PFC) is a fast acting mechanism used to ensure high-quality power for the grid that is becoming an increasingly attractive option for load participation. Due to speed requirement and other considerations, it is often desirable to have distributed control laws. Current distributed PFC designs assume that the costs at each geographic location are independent. However, many networked systems, such as those for cloud computing, have interdependent costs across locations and therefore need geographic coordination. In this paper, distributed control laws are designed for interdependent, geo-distributed loads in PFC based on the optimality conditions of the global system. The controlled frequencies are provably stable, and the final equilibrium point is proven to strike an optimal balance between load participation and the frequency's deviation from its nominal set point. The case with a significant communication delay for control is also analyzed. We evaluate the proposed control laws with realistic numerical simulations. Under current technology, the proposed control laws achieve a convergence time that is smaller than droop control alone and is comparable to that of distributed control without interdependent costs. Results also highlight significant cost savings over existing approaches under a variety of settings.

math.OC↗