An official website of the United States government
Transient computing has become popular in public cloud environments for running delay-insensitive batch and data processing applications at low cost. Since transient cloud servers can be revoked at any time by the cloud provider, they are considered unsuitable for running interactive application such as web services. In this paper, we present VM deflation as an alternative mechanism to server preemption for reclaiming resources from transient cloud servers under resource pressure. Using real traces from top-tier cloud providers, we show the feasibility of using VM deflation as a resource reclamation mechanism for interactive applications in public clouds. We show how current hypervisor mechanisms can be used to implement VM deflation and present cluster deflation policies for resource management of transient and on-demand cloud VMs. Experimental evaluation of our deflation system on a Linux cluster shows that microservice-based applications can be deflated by up to 50% with negligible performance overhead. Our cluster-level deflation policies allow overcommitment levels as high as 50%, with less than a 1% decrease in application throughput, and can enable cloud platforms to increase revenue by 30%
more »
« less
Resource Deflation: A New Approach For Transient Resource Reclamation
Data centers and clouds are increasingly offering low-cost computational resources in the form of transient virtual machines. Whenever demand for computational resources exceeds their availability, transient resources can reclaimed by preempting the transient VMs. Conventionally, these transient VMs are used by low-priority applications that can tolerate the disruption caused by preemptions. In this paper we propose an alternative approach for reclaiming resources, called resource deflation. Resource deflation allows applications to dynamically shrink (and expand) in response to resource pressure, instead of being preempted outright. Deflatable VMs allow applications to continue running even under resource pressure, and increase the utility of low-priority transient resources. Deflation uses a dynamic, multi-level cascading reclamation technique that allows applications, operating systems, and hypervisors to implement their own policies for handling resource pressure. For distributed data processing, machine learning, and deep neural network training, our multi-level approach reduces the performance degradation by up to 2x compared to existing preemption-based approaches. When deflatable VMs are deployed on a cluster, our policies allow up to 1.6x utilization without the risk of preemption.
more »
« less
- Award ID(s):
- 1802523
- PAR ID:
- 10092716
- Date Published:
- Journal Name:
- Proceedings of the Fourteenth EuroSys Conference
- Page Range / eLocation ID:
- 1 to 17
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
In multi-agent systems, limited resources must be shared by individuals during missions to maximize the group utility of the system in the field. In this paper, we present a generalized adaptive self-organization process for multi-agent systems featuring fast and efficient distribution of a consumable and refillable on-board resource throughout the group. An adaptive inter-agent spacing (AIS) controller based on individual resource levels is proposed that spaces out high resource bearing agents throughout the group including the group boundary extrema, and allows low resource bearing agents to adaptively occupy the in-between spaces receiving resource from the high resource bearing agents without over-crowding. Experimental results for cases with and without the proposed AIS controller validate faster convergence of individual resource levels to the group mean resource level using the proposed AIS controller. The generalized approach of the self-organizing process allows flexibility in adapting the proposed AIS controller for various multi-agent applications.more » « less
-
Preemptive scheduling policies, which allow pausing jobs mid-service, are ubiquitous because they allow important jobs to receive service ahead of unimportant jobs that would otherwise delay their completion. The canonical example is Shortest Remaining Processing Time (SRPT), which preemptively serves the job with least remaining work at every moment in time [9]. There is a robust literature analyzing response time (elapsed time between a job's arrival and completion) in the M/G/1 queue under many preemptive policies [6, 10, 11], shedding light on questions such as how preemption affects the mean and tail of response time, and whether preemption is unfair towards low-priority jobs.more » « less
-
We present Memtrade, the first practical marketplace for disaggregated memory clouds. Clouds introduce a set of unique challenges for resource disaggregation across different tenants, including resource harvesting, isolation, and matching. Memtrade allows producer virtual machines (VMs) to lease both their unallocated memory and allocated-but-idle application memory to remote consumer VMs for a limited period of time. Memtrade does not require any modifications to host-level system software or support from the cloud provider. It harvests producer memory using an application-aware control loop to form a distributed transient remote memory pool with minimal performance impact; it employs a broker to match producers with consumers while satisfying performance constraints; and it exposes the matched memory to consumers through different abstractions. As a proof of concept, we propose two such memory access interfaces for Memtrade consumers -- a transient KV cache for specified applications and a swap interface that is application-transparent. Our evaluation using real-world cluster traces shows that Memtrade provides significant performance benefit for consumers (improving average read latency up to 2.8X) while preserving confidentiality and integrity, with little impact on producer applications (degrading performance by less than 2.1%).more » « less
-
Several recent studies have investigated the virtual machine (VM) provisioning problem for requests with time constraints (deadlines) in cloud systems. These studies typically assumed that a request is associated with a single execution time when running on VMs with a given resource demand. In this paper, we consider modern applications that are normally implemented with generic frameworks that allow them to execute with various numbers of threads on VMs with different resource demands. For such applications, it is possible for the users to specify multiple execution options (MEOs) for a request where each execution option is represented by a certain number of VMs with some resources to run the application and its corresponding execution time. We investigate the problem of virtual machine provisioning for such time-sensitive requests with MEOs in resource-constrained clouds. By incorporating the MEOs of requests, we propose several novel and flexible VM provisioning schemes that carefully balance resource usage efficiency, input workloads and request deadlines with the objective of achieving higher resource utilization and system benefits. We evaluated the proposed MEO-aware schemes on various workloads with both benchmark requests and synthetic requests. The results show that our MEO-aware algorithms outperform the state-of-the-art schemes that consider only a single execution option of requests by serving up to 38% more requests and achieving up to 27% more benefits.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3302424.3303945
