ETICA: <u>E</u>fficient <u>T</u>wo-Level <u>I</u>/O <u>C</u>aching <u>A</u>rchitecture for Virtualized Platforms

نویسندگان

چکیده

In this paper, we propose an Efficient Two-Level I/O Caching Architecture (ETICA) for virtualized platforms that can significantly improve latency, endurance, and cost (in terms of cache size) while preserving the reliability write-pending data blocks. As opposed to previous one-level caching schemes in platforms, our proposed architecture 1) provides two levels by employing both Dynamic Random-Access Memory (DRAM) SSD layer 2) effectively partitions space between running VMs achieve maximum performance minimum size. To manage two-level cache, unlike reuse distance calculation such as Useful Reuse Distance (URD), which only consider request type neglect impact write policy, a new metric, Policy Optimized (POD). The key idea POD is calculate estimate amount DRAM+SSD allocate considering policy. Doing so results enhanced reduced size due allocation blocks requests would be served cache. ETICA maintains improves assigning effective fixed policy at each level hierarchy promotion eviction methods levels. Our extensive experiments conducted with real implementation storage show 45% higher performance, compared state-of-the-art improving endurance 51.7% 33.8%, respectively.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Resource allocation algorithms for virtualized service hosting platforms

Commodity clusters are used routinely for deploying service hosting platforms. Due to hardware and operation costs, clusters need to be shared among multiple services. Crucial for enabling such shared hosting platforms is virtual machine (VM) technology, which allows consolidation of hardware resources. A key challenge, however, is to make appropriate decisions when allocating hardware resource...

متن کامل

Trusted Integrity Measurement and Reporting for Virtualized Platforms

Verifiable trust is a desirable property for computing platforms. Current trusted computing systems developed by Trusted Computing Group (TCG) provide verifiable trust by taking immutable snapshots of the whole set of platform components. It is, however, difficult to use this technology directly in virtualized platforms because of complexity and dynamic changes of platform components. In this p...

متن کامل

An Energy-efficient Task Scheduler in Virtualized Cloud Platforms

In cloud platforms, virtualization technology has been widely applied for deploying largescale IT-infrastructures due to its flexibility and extendibility. However, the extra software layer introduced by virtualization technology also raises many performance issues. One of them is the energy-efficiency losses when massive I/O-intensive tasks are running on virtualized servers. In this paper, we...

متن کامل

Intermediate Level Components for Reconfigurable Platforms

This paper addresses the problem of development productivity on reconfigurable platforms. Due to the availability of generic low level tools and powerful logic synthesis tools, it becomes possible to define portable components that have both a high level behavior and attributes for physical synthesis. The behavior of a component can be fixed at compile time using concise specifications that wil...

متن کامل

TCG Based Approach for Secure Management of Virtualized Platforms State - of - the - art ( June 05 , 2010 )

There is a strong trend shift in the favor of adopting virtualization to get business benefits. The provisioning of virtualized enterprise resources is one kind of many possible scenarios. Where virtualization promises clear advantages it also poses new security challenges which need to be addressed to gain stakeholders confidence in the dynamics of new environment. One important facet of these...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE Transactions on Parallel and Distributed Systems

سال: 2021

ISSN: ['1045-9219', '1558-2183', '2161-9883']

DOI: https://doi.org/10.1109/tpds.2021.3066308