Reducing the number of response time service level objective violations by a cloud‐HPC convergence scheduler

Abstract

Summary Job scheduling is an old topic in High‐Performance Computing (HPC), and it is more and more studied in data centers. Large data centers are often split into separate partitions for cloud computing and HPC; each partition normally has its specific scheduler. The possibility of migrating jobs from the HPC partition to the cloud one is a topic widely discussed in the literature. However, job migration from cloud to HPC is a much less explored topic. Nevertheless, such migration may be useful in many situations, in particular when the HPC platform has a low resource usage level, and the cloud usage level is high. A large number of jobs that could migrate from the cloud to the HPC partition may be observed in Google data center workloads. Job scheduling using overbooking strategy is seen as the main reason for the high resource usage level in clouds. However, overbooking can lead to a high rate of rescheduling and job dumping, which potentially causes response time violations. This work shows that HPC platforms can host and execute some cloud jobs with low interference in HPC jobs and a low number of response time violations. We introduce the definition of a cloud‐HPC convergence area and propose a job scheduling strategy for it, aiming at reducing the number of response time violations of cloud jobs without interfering with HPC jobs execution. Our proposal is formally defined and then evaluated in different execution scenarios, using the SimGrid simulation framework, with workload data from production HPC grid. The experimental results show that often, there is a large number of empty areas in the scheduling plan of HPC platforms, which makes it possible to allocate cloud jobs by backfilling. This is due to the sparse HPC job submission pattern and the low resource usage level in some HPC platforms. One performed simulation scenario considered a set of 11K parallel HPC jobs running on a 2560‐processor platform having an average resource usage level of 38.0%. The proposed convergence scheduler succeeded to inject around 267K cloud jobs in the HPC platform, with a response time violation rate under 0.00094% for such jobs, considering 80 processors in the convergence area and no effects on the HPC workload. This experiment considered cloud jobs based on job features of Google public cloud workloads, with a processing time slack factor of 1.25 (which is considered as high priority in the Google cloud SLA—Service Level Agreement). Usually, most cloud jobs show a slack factor higher than 1.25 (most cloud jobs are medium or low priority). The same simulation, repeated with a higher slack factor (4), showed no response time violations.

Efficient HPC and Energy-Aware Proactive Dynamic VM Consolidation in Cloud Computing

“…However, the paper neither takes into account the HPC workload nor discusses the issues of cross-application interference. Kraemer et al [19] developed a job migration mechanism for transferring jobs from the cloud environment to the high-performance computing environment. The primary goal is to reduce the amount of response time violations associated with cloud jobs while not interfering with the execution of HPC jobs.…”

An efficient virtual machine allocation algorithm for parallel and distributed simulation applications

“…Apart from them, some other researchers put forward resource allocation algorithms based on cost or profit. 38 However, these methods mainly solve the problems of assigning tasks to VMs or allocating VMs to hosts. The former mode tries to determine VMs on which the applications will be executed before the application deployment according to the resource usage in a period.…”

“…In parallel applications, batch scheduling was often adopted to organize the execution parallel application. 38 However, these methods mainly solve the problems of assigning tasks to VMs or allocating VMs to hosts. The precondition of these approaches is that an application can be completed by a single VM.…”

Cloud computing and big data: Technologies and applications

“…Results also showed that the selected value should be above a particular threshold as unnecessary frequent checkpointing may lead to performance degradation. Kraemer et al, in their article “ Reducing the number of response time Service Level Objective violations by a cloud‐HPC convergence scheduler ,” introduced a scheduling strategy for migrating jobs from the cloud environment to the HPC environment. The main aim is to reduce the number of response time violations of cloud jobs without interfering with HPC job execution.…”