A Performance Study on the VM Startup Time in the Cloud

@article{Mao2012APS,
  title={A Performance Study on the VM Startup Time in the Cloud},
  author={Ming Mao and Marty Humphrey},
  journal={2012 IEEE Fifth International Conference on Cloud Computing},
  year={2012},
  pages={423-430}
}
  • Ming Mao, M. Humphrey
  • Published 24 June 2012
  • Computer Science
  • 2012 IEEE Fifth International Conference on Cloud Computing
One of many advantages of the cloud is the elasticity, the ability to dynamically acquire or release computing resources in response to demand. However, this elasticity is only meaningful to the cloud users when the acquired Virtual Machines (VMs) can be provisioned in time and be ready to use within the user expectation. The long unexpected VM startup time could result in resource under-provisioning, which will inevitably hurt the application performance. A better understanding of the VM… 
Benchmarking VM Startup Time in the Cloud
TLDR
This paper benchmark VM startup time in Amazon EC2, Microsoft Azure and Google Cloud for factors like instance type, time of the day, instance location, cluster creation and cluster resize.
An Empirical Analysis of VM Startup Times in Public IaaS Clouds
TLDR
This work performs a thorough analysis of VM startup times in two IaaS clouds (AWS, GCP), collecting 300,000 data points from each provider by applying diverse VM configurations to find several important factors that can change VM startup time significantly.
An Approach for Dynamic Scaling of Resources in Enterprise Cloud
  • K. KanagalaK. Sekaran
  • Computer Science
    2013 IEEE 5th International Conference on Cloud Computing Technology and Science
  • 2013
TLDR
This work developed an approach where the thresholds adapt in advance and these thresholds are functions of load trend, which makes VM readily available when needed.
Prebaked µVMs: Scalable, Instant VM Startup for IaaS Clouds
TLDR
This work proposes using so-called prebaked uVMs, snapshots of minimal VMs that can be quickly resumed and then configured to application needs by hot-plugging resources to speed up virtual machine start up.
Fast and scalable virtual machine deployment
TLDR
It is observed that only a tiny part of the VM image is needed for the VM to be able to start up, and it is proposed using small caches for VM images to overcome the VM startup bottlenecks.
Cost-Effective Virtual Machine Image Replication Management for Cloud Data Centers
  • Dian ShenFang DongJunxue ZhangJunzhou Luo
  • Computer Science
    2014 IEEE Intl Conf on High Performance Computing and Communications, 2014 IEEE 6th Intl Symp on Cyberspace Safety and Security, 2014 IEEE 11th Intl Conf on Embedded Software and Syst (HPCC,CSS,ICESS)
  • 2014
TLDR
A stochastic model based on queueing theory is proposed, which captures the main factors in image provisioning to optimize the number and placement of image replication, so as to manage the VM images in a cost-effective manner.
Reducing VM Startup Time and Storage Costs by VM Image Content Consolidation
TLDR
A novel approach for consolidating size and contents of VMIs is developed and compared with an unmodified ConPaaS VMI, an open-source Platform-as-a-Service runtime, results in up to four times reduction of the disk size, three times speedup for the VM startup time, and three times reduction in storage cost.
VMThunder: Fast Provisioning of Large-Scale Virtual Machine Clusters
TLDR
VMThunder, a new VM provisioning tool, which downloads data blocks on demand during the VM booting process and speeds up VM image streaming by strategically integrating peer-to-peer streaming techniques with enhanced optimization schemes such as transfer on demand, cache on read, snapshot on local, and relay on cache.
Effects of Reducing VMs Management Times on Elastic Applications
TLDR
This work evaluates to what extent the reduction of the time required by VM management operations, namely deployment and shutdown, impacts the performance of applications and the behavior of reactive and proactive auto-scaling policies.
A Reference Model for Virtual Machine Launching Overhead
TLDR
Efficient resource allocation algorithms can be developed for cloud bursting process to minimize the operational cost and resource waste with the guidance of the developed reference model.
...
...

References

SHOWING 1-10 OF 19 REFERENCES
Twinkle: A fast resource provisioning mechanism for internet services
TLDR
This paper develops a fast start technique for cloud applications by restoring previously created VM snapshots of fully initialized application, and proposes a set of optimizations that allow an application to start running with only partially loaded memory, yet without noticeable performance penalty during its subsequent execution.
Jump-Start Cloud: Efficient Deployment Framework for Large-Scale Cloud Applications
TLDR
This work designs jump-start cloud, under which an efficient cloud deployment scheme is proposed for minimizing cloud time, and shows that the design saves application installation time and meanwhile, keeps application running time reasonably low, thus makes cloud time shorter.
Debunking Real-Time Pricing in Cloud Computing
  • S. Wee
  • Computer Science
    2011 11th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing
  • 2011
TLDR
It is concluded that the observed monetary incentive is not large enough to motivate users to shift their workloads in real-time pricing, and both average cost savings and price fluctuation have not been meaningfully changed over time.
VDN: Virtual machine image distribution network for cloud data centers
TLDR
A chunk-level Virtual machine image Distribution Network (VDN) is proposed, which takes advantage of the hierarchical network topology of data centers to reduce the VM instance provisioning time and also to minimize the overhead of maintaining chunk location information.
Early observations on the performance of Windows Azure
TLDR
In general, good performance of the Windows Azure mechanisms are observed, although the average 10 minute VM startup time and the worst-case 2x slowdown for SQL Azure in certain situations must be accounted for in application design.
Auto-scaling to minimize cost and meet application deadlines in cloud workflows
  • Ming MaoM. Humphrey
  • Computer Science
    2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)
  • 2011
TLDR
This paper presents an approach whereby the basic computing elements are virtual machines (VMs) of various sizes/costs, jobs are specified as workflows, users specify performance requirements by assigning (soft) deadlines to jobs, and the goal is to ensure all jobs are finished within their deadlines at minimum financial cost.
A Performance Analysis of EC2 Cloud Computing Services for Scientific Computing
TLDR
This work analyzes the performance of the Amazon EC2 platform using micro-benchmarks and kernels and indicates that the current cloud services need an order of magnitude in performance improvement to be useful to the scientific community.
Managing Peak Loads by Leasing Cloud Infrastructure Services from a Spot Market
TLDR
This paper defines different provisioning policies and investigates the use of spot instances compared to normal instances in terms of cost savings and total breach time of tasks in the queue.
Reducing Costs of Spot Instances via Checkpointing in the Amazon Elastic Compute Cloud
TLDR
Based on the real price history of EC2 spot instances, this work compares several adaptive check pointing schemes in terms of monetary costs and improvement of job completion times and shows that its approach can reduce significantly both price and the task completion times.
FVD: A High-Performance Virtual Machine Image Format for Cloud
TLDR
Fast Virtual Disk is a new virtual machine (VM) image format and the corresponding block device driver developed for QEMU that enables instant VM creation and instant VM migration, even if the VM image is stored on direct-attached storage.
...
...