Improving Agility and Elasticity in Bare-metal Clouds

Omote, Yushi; Shinagawa, Takahiro; Kato, K.

doi:10.1145/2694344.2694349

Cited by 21 publications

(9 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Today's bare-metal clouds take many tens of minutes to allocate and provision a server [62]. Further, scrubbing the disk can take many hours; an operation required for stateful bare metal clouds whenever a server is being transferred between one tenant and another.…”

Section: Elasticitymentioning

confidence: 99%

Supporting Security Sensitive Tenants in a Bare-Metal Cloud

Mosayyebzadeh¹,

Mohan²,

Tikale³

et al. 2019

Preprint

View full text Add to dashboard Cite

Bolted is a new architecture for bare-metal clouds that enables tenants to control tradeoffs between security, price, and performance. Security-sensitive tenants can minimize their trust in the public cloud provider and achieve similar levels of security and control that they can obtain in their own private data centers. At the same time, Bolted neither imposes overhead on tenants that are security insensitive nor compromises the flexibility or operational efficiency of the provider. Our prototype exploits a novel provisioning system and specialized firmware to enable elasticity similar to virtualized clouds. Experimentally we quantify the cost of different levels of security for a variety of workloads and demonstrate the value of giving control to the tenant. * Mosayyebzadeh, Mohan, and Tikale contributed equally. † DISTRIBUTION STATEMENT A. Approved for public release. Distribution is unlimited.

show abstract

Section: Elasticitymentioning

confidence: 99%

Supporting Security Sensitive Tenants in a Bare-Metal Cloud

Mosayyebzadeh¹,

Mohan²,

Tikale³

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…The hypervisor is implemented based on BitVisor [3], which is a lightweight bare-metal hypervisor. BitVisor is opensource and has been used in several studies [1,2]. BitVisor does not require a host OS and starts execution as part of the boot process in the bootloader.…”

Section: Hypervisormentioning

confidence: 99%

“…We have implemented FaultVisor based on BitVisor [3], which is a versatile platform for mediating device access for various purposes [1,2]. In our evaluation, FaultVisor and a manual analysis based on the test results identified 41 fault patterns that disclose problems.…”

Section: Introductionmentioning

confidence: 99%

Testing device drivers against hardware failures in real environments

Takekoshi

Shinagawa

Kato

2016

Proceedings of the 31st Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

Hardware failures are a major cause of computer system crashes. To improve system reliability, making device drivers sufficiently robust to tolerate hardware failures is important. Unfortunately, testing device drivers against hardware failures is not easy because hardware failures must be artificially produced in a controlled manner. Prior work has required source code, OS-dependent implementation efforts, or special virtual environments for testing in which it is difficult to find problems that only occur in real environments. This paper presents FaultVisor, a bare-metal hypervisor specialized for testing device drivers. FaultVisor injects pseudo hardware failures by modifying access to real hardware using the hypervisor layer, allowing closed-source and OS-independent driver testing in real environments, and easy insertion of test systems into existing systems. In our evaluation, FaultVisor discovered 41 fault patterns that disclose problems in device drivers, 30 of which led to critical system failures.

show abstract

“…A number of industry and research projects have attacked the performance and functionality challenges of provisioning bare-metal nodes [15], [16], [17], [18], [19], [20]; automating the bare-metal provisioning process, reducing the management overhead of the cloud provider, and improving the performance of copying the image to the server's disk. For example, Omote et al [20] proposed a lazy copy approach that copies the OS image in the background after the operating system is booted using a remote disk.…”

Section: Introductionmentioning

confidence: 99%

M2: Malleable Metal as a Service

Mohan

Türk

Gudimetla

et al. 2018

2018 IEEE International Conference on Cloud Engineering (IC2E)

View full text Add to dashboard Cite

Existing bare-metal cloud services that provide users with physical nodes have a number of serious disadvantage over their virtual alternatives, including slow provisioning times, difficulty for users to release nodes and then reuse them to handle changes in demand, and poor tolerance to failures. We introduce M2, a bare-metal cloud service that uses network-mounted boot drives to overcome these disadvantages. We describe the architecture and implementation of M2 and compare its agility, scalability and performance to existing systems. We show that M2 can reduce provisioning time by over 50% while offering richer functionality, and comparable run time performance with respect to tools that provision images into local disks. M2 is open source and available at https://github.com/CCI-MOC/ims. 1 Previous provisioning systems exploited remote storage in special purpose environments, like HPC clusters, where all nodes boot the same kernel.2 As the focus of this work is to improve the provisioning time M2 only network-mounts boot drives that hosts the OS and applications. Data drives are still hosted on the local disks.

show abstract

Improving Agility and Elasticity in Bare-metal Clouds

Cited by 21 publications

References 27 publications

Supporting Security Sensitive Tenants in a Bare-Metal Cloud

Supporting Security Sensitive Tenants in a Bare-Metal Cloud

Testing device drivers against hardware failures in real environments

M2: Malleable Metal as a Service

Contact Info

Product

Resources

About