skip to main content


Title: Autoscaling High-Throughput Workloads on Container Orchestrators
High-throughput computing (HTC) workloads seek to complete as many jobs as possible over a long period of time. Such workloads require efficient execution of many parallel jobs and can occupy a large number of resources for a longtime. As a result, full utilization is the normal state of an HTC facility. The widespread use of container orchestrators eases the deployment of HTC frameworks across different platforms,which also provides an opportunity to scale up HTC workloads with almost infinite resources on the public cloud. However, the autoscaling mechanisms of container orchestrators are primarily designed to support latency-sensitive microservices, and result in unexpected behavior when presented with HTC workloads. In this paper, we design a feedback autoscaler, High Throughput Autoscaler (HTA), that leverages the unique characteristics ofthe HTC workload to autoscales the resource pools used by HTC workloads on container orchestrators. HTA takes into account a reference input, the real-time status of the jobs’ queue, as well as two feedback inputs, resource consumption of jobs, and the resource initialization time of the container orchestrator. We implement HTA using the Makeflow workload manager, WorkQueue job scheduler, and the Kubernetes cluster manager. We evaluate its performance on both CPU-bound and IO-bound workloads. The evaluation results show that, by using HTA, we improve resource utilization by 5.6×with a slight increase in execution time (about 15%) for a CPU-bound workload, and shorten the workload execution time by up to 3.65×for an IO-bound workload.  more » « less
Award ID(s):
1642409
NSF-PAR ID:
10210595
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
IEEE Conference on Cluster Computing
Page Range / eLocation ID:
142 to 152
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Serverless computing allows customers to submit their jobs to the cloud for execution, with the resource provisioning being taken care of by the cloud provider. Serverless functions are often short-lived and have modest resource requirements, thereby presenting an opportunity to improve server utilization by colocating with latency-sensitive customer workloads. This paper presents ServerMore, a server-level resource manager that opportunistically colocates customer serverless jobs with serverful customer VMs. ServerMore dynamically regulates the CPU, memory bandwidth, and LLC resources on the server to ensure that the colocation between serverful and serverless workloads does not impact application tail latencies. By selectively admitting serverless functions and inferring the performance of black-box serverful workloads, ServerMore improves resource utilization on average by 35.9% to 245% compared to prior works; while having a minimal impact on the latency of both serverful applications and serverless functions. 
    more » « less
  2. High performance computing (HPC) is undergoing significant changes. The emerging HPC applications comprise both compute- and data-intensive applications. To meet the intense I/O demand from emerging data-intensive applications, burst buffers are deployed in production systems. Existing HPC schedulers are mainly CPU-centric. The extreme heterogeneity of hardware devices, combined with workload changes, forces the schedulers to consider multiple resources (e.g., burst buffers) beyond CPUs, in decision making. In this study, we present a multi-resource scheduling scheme named BBSched that schedules user jobs based on not only their CPU requirements, but also other schedulable resources such as burst buffer. BBSched formulates the scheduling problem into a multi-objective optimization (MOO) problem and rapidly solves the problem using a multi-objective genetic algorithm. The multiple solutions generated by BBSched enables system managers to explore potential tradeoffs among various resources, and therefore obtains better utilization of all the resources. The trace-driven simulations with real system workloads demonstrate that BBSched improves scheduling performance by up to 41% compared to existing methods, indicating that explicitly optimizing multiple resources beyond CPUs is essential for HPC scheduling. 
    more » « less
  3. Traditionally, HPC workloads have been deployed in bare-metal clusters; but the advances in virtualization have led the pathway for these workloads to be deployed in virtualized clusters. However, HPC cluster administrators/providers still face challenges in terms of resource elasticity and virtual machine (VM) provisioning at large-scale, due to the lack of coordination between a traditional HPC scheduler and the VM hypervisor (resource management layer). This lack of interaction leads to low cluster utilization and job completion throughput. Furthermore, the VM provisioning delays directly impact the overall performance of jobs in the cluster. Hence, there is a need for effectively provisioning virtualized HPC clusters, which can best-utilize the physical hardware with minimal provisioning overheads.Towards this, we propose Multiverse, a VM provisioning framework, which can dynamically spawn VMs for incoming jobs in a virtualized HPC cluster, by integrating the HPC scheduler along with VM resource manager. We have implemented this framework on the Slurm scheduler along with the vSphere VM resource manager. In order to reduce the VM provisioning overheads, we use instant cloning which shares both the disk and memory with the parent VM, when compared to full VM cloning which has to boot-up a new VM from scratch. Measurements with real-world HPC workloads demonstrate that, instant cloning is 2.5× faster than full cloning in terms of VM provisioning time. Further, it improves resource utilization by up to 40%, and cluster throughput by up to 1.5×, when compared to full clone for bursty job arrival scenarios. 
    more » « less
  4. null (Ed.)
    Fast networks and the desire for high resource utilization in data centers and the cloud have driven disaggregation. Application compute is separated from storage, but this leads to high overheads when data must move over the network for simple operations on it. Alternatively, systems could allow applications to run application logic within storage via user-defined functions. Unfortunately, this ties provisioning and utilization of storage and compute resources together again. We present a new approach to executing storage-level functions in an in-memory key-value store that avoids this problem by dynamically deciding where to execute functions over data. Users write storage functions that are logically decoupled from storage, but storage servers choose where to run invocations of these functions physically. By using a server-internal cost model and observing function execution, servers choose to directly run inexpensive functions, while preferring to execute functions with high CPU-cost at client machines. We show that with this approach storage servers can reduce network request processing costs, avoid server compute bottlenecks, and improve aggregate storage system throughput. We realize our approach on an in-memory key-value store that executes 3.2 million strict serializable user-defined storage functions per second with 100 us response times. When running a mix of logic from different applications, it provides throughput better than running that logic purely at storage servers (85% more) or purely at clients (10% more). For our workloads, it also reduces latency (up to 2x) and transactional aborts (up to 33%) over pure client-side execution. 
    more » « less
  5. Finding from a big graph those subgraphs that satisfy certain conditions is useful in many applications such as community detection and subgraph matching. These problems have a high time complexity, but existing systems that attempt to scale them are all IO-bound in execution. We propose the first truly CPU-bound distributed framework called G-thinker for subgraph finding algorithms, which adopts a task-based computation model, and which also provides a user-friendly subgraph-centric vertex-pulling API for writing distributed subgraph finding algorithms that can be easily adapted from existing serial algorithms. To utilize all CPU cores of a cluster, G-thinker features (1) a highly concurrent vertex cache for parallel task access and (2) a lightweight task scheduling approach that ensures high task throughput. These designs well overlap communication with computation to minimize the idle time of CPU cores. To further improve load balancing on graphs where the workloads of individual tasks can be drastically different due to biased graph density distribution, we propose to prioritize the scheduling of those tasks that tend to be long running for processing and decomposition, plus a timeout mechanism for task decomposition to prevent long-running straggler tasks. The idea has been integrated into a novelty algorithm for maximum clique finding (MCF) that adopts a hybrid task decomposition strategy, which significantly improves the running time of MCF on dense and large graphs: The algorithm finds a maximum clique of size 1,109 on a large and dense WikiLinks graph dataset in 70 minutes. Extensive experiments demonstrate that G-thinker achieves orders of magnitude speedup compared even with the fastest existing subgraph-centric system, and it scales well to much larger and denser real network data. G-thinker is open-sourced at http://bit.ly/gthinker with detailed documentation. 
    more » « less