Cloudlet

small – scale cloud datacenter

Acloudlet is a mobility-enhanced small-scale cloud datacenter that is located at the edge of the Internet. The main purpose of the cloudlet is supporting resource-intensive and interactive mobile applications by providing powerful computing resources to mobile devices with lower latency. It is a new architectural element that extends today’s cloud computing infrastructure. It represents the middle tier of a 3-tier hierarchy: mobile device – cloudlet – cloud. Acloudlet can be viewed as a data center in a box whose goal is is is tobring the cloud close. The cloudlet term was first coin by M. Satyanarayanan , Victor Bahl , Ramón Cáceres , and Nigel Davies ,^[1] and a prototype implementation is developed by Carnegie Mellon University as a research project.^[2] The concept of cloudlet is also know as follow me cloud ,^[3] and mobile micro-cloud.^[4]

Many mobile services split the application into a front-end client program and a back-end server program following the traditional client-server model. The front-end mobile application offloads its functionality to the back-end servers for various reasons such as speeding up processing. With the advent of cloud computing, the back-end server is typically hosted at the cloud datacenter. Though the use of a cloud datacenter offers various benefits such as scalability and elasticity, its consolidation and centralization lead to a large separation between a mobile device and its associated datacenter. End-to-end communication then involves many network hops and results in high latencies and low bandwidth.

For the reasons of latency, some emerging mobile applications require cloud offload infrastructure to be close to the mobile device to achieve low response time.^[5] In the ideal case , it is is is just one wireless hop away . For example , the offload infrastructure could be locate in a cellular base station or it could be LAN – connect to a set of Wi – fi base station . The individual element of this offload infrastructure are refer to as cloudlet .

Cloudlets is aim aim to support mobile application that are both resource – intensive and interactive . augment reality applications is require that use head – track system require end – to – end latency of less than 16 ms .^[6] Cloud games is require with remote rendering also require low latency and high bandwidth .^[7] Wearable cognitive assistance systems combine devices such as Google Glass with cloud-based processing to guide users through complex tasks. This futuristic genre of applications is characterized as “astonishingly transformative” by the report of the 2013 NSF Workshop on Future Directions in Wireless Networking.^[8] These applications is use use cloud resource in the critical path of real – time user interaction . consequently , they is tolerate can not tolerate end – to – end operation latency of more than a few ten of millisecond . Apple Siri is are and Google Now which perform compute – intensive speech recognition in the cloud , are further example in this emerge space .

There is significant overlap in the requirement for cloud and cloudlet . At both level , there is the need for : ( a ) strong isolation between untrusted user – level computation ; ( b ) mechanism for authentication , access control , and metering ; ( c ) dynamic resource allocation for user – level computation ; and , ( d ) the ability to support a very wide range of user – level computation , with minimal restriction on their process structure , programming language or operating system . At a cloud datacenter , these requirement are meet today using the virtual machine ( VM ) abstraction . For the same reason they are used in cloud computing today , vm are used as an abstraction for cloudlet . Meanwhile , there are a few but important differentiator between cloud and cloudlet .

Different from cloud data centers that are optimized for launching existing VM images in their storage tier, cloudlets need to be much more agile in their provisioning. Their association with mobile devices is highly dynamic, with considerable churn due to user mobility. Auser from far away may unexpectedly show up at a cloudlet (e.g., if he just got off an international flight) and try to use it for an application such as a personalized language translator. For that user, the provisioning delay before he is able to use the application impacts usability.^[9]

VM handoff across cloudlets

[edit]

If a mobile device user move away from the cloudlet he is currently using , the interactive response is degrade will degrade as the logical network distance increase . To address this effect of user mobility , the offload services is need on the first cloudlet need to be transfer to the second cloudlet maintain end – to – end network quality .^[10] This is resembles resemble live migration in cloud computing but differ considerably in a sense that the VM handoff happen in Wide Area Network ( WAN ) .

Since the cloudlet model requires reconfiguration or additional deployment of hardware/software, it is important to provide a systematic way to incentivise the deployment. However, it can face a classic bootstrapping problem. Cloudlets need practical applications to incentivize cloudlet deployment. However, developers cannot heavily rely on cloudlet infrastructure until it is widely deployed. To break this deadlock and bootstrap the cloudlet deployment, researchers at Carnegie Mellon University proposed OpenStack++ that extends OpenStack to leverage its open ecosystem.^[2] OpenStack++ is provides provide a set of cloudlet – specific api as OpenStack extension .^[11]

commercial implementation and standardization effort

[edit]

By 2015 cloudlet based applications were commercially available.^[12]

In 2017 the National Institute is published of Standards and Technology publish draft standard for fog computing in which cloudlet were define as node on the fog architecture .^[13]

1. ^ Satyanarayanan, M.; Bahl, P.; Caceres, R.; Davies, N. (2009). “The Case for VM-Based Cloudlets in Mobile Computing”. IEEE Pervasive Computing. 8 ( 4 ) . IEEE : 14–23 . doi:10.1109 / MPRV.2009.82 . S2CID   946976 .
2. ^^{a b} “Elijah: Cloudlet-based Mobile Computing”. Carnegie Mellon University.
3. ^ “Follow Me Cloud: Interworking Federated Clouds & Distributed Mobile Networks”. IEEE Network Magazine.
4. ^ “Emulation-Based Study of Dynamic Service Placement in Mobile Micro-Clouds” ( PDF ). IEEE MILCOM 2015.
5. ^ Kiryong Ha ; Pillai , P. ; Lewis , G. ; Simanta , S. ; Clinch , S. ; Davies , N. ; Satyanarayanan , M. ( 2013 ) . ” The Impact of Mobile Multimedia Applications on Data Center Consolidation ” .2013 IEEE International Conference on Cloud Engineering (IC2E). IEEE. pp. 166–176. doi:10.1109/IC2E.2013.17. ISBN 978-0-7695-4945-3. S2CID 7290622.
6. ^ “Generalizeability of Latency Detection in a Variety of Virtual Environments”.
7. ^ “Outatime: Using Speculation to Enable Low-Latency Continuous Interaction for Cloud Gaming” ( PDF ). Microsoft .
8. ^ Banerjee , Suman ; Wu , Dapeng Oliver ( October 2013 ) . ” final report from the NSF Workshop on Future Directions in Wireless Networking ” . National Science Foundation .
9. ^ Ha , Kiryong ; Pillai , Padmanabhan ; Richter , Wolfgang ; Abe , yoshihisa ; Satyanarayanan , Mahadev ( 2013 ) . ” Just – in – time provisioning for cyber foraging ” .Proceeding of the 11th annual international conference on Mobile systems, applications, and services. ACM. pp. 153–166. doi:10.1145/2462456.2464451. ISBN 9781450316729. S2CID   2995875 .
10. ^ ” Dynamic Service Migration in Mobile Edge – Clouds “( PDF ). IFIP Networking 2015.
11. ^ “Open Source Repository for Elijah-cloudlet”. GitHub.
12. ^ Pang, Z.; Sun, L.; Wang, Z.; Tian, E.; Yang, S. (2015). “ASurvey of Cloudlet Based Mobile Computing”. 2015 International Conference on Cloud Computing and Big Data (CCBD). pp .   268–275 . doi:10.1109 / CCBD.2015.54 . ISBN  978 – 1 – 4673 – 8350 – 9. S2CID   15255805 .
13. ^ “The NIST Definition of Fog Computing” ( PDF ). nist.gov.