Importance of Application-Level Resource Management in Multi-Cloud Deployments

Cloud service providers started with Infrastructure as a Service (IaaS) offerings and over time expanded into Platform as a Service (PaaS) and Software as a Service (SaaS). Even though each provider has a rich product offering, there are many scenarios where a multi-cloud strategy is desirable: utilizing economic dynamics, preventing data lock-in with one vendor, circumventing geographic restrictions, complying with local regulations, or combining on-premise and public-cloud resources. The challenge from a consumer perspective with multi-cloud deployments is the lack of a common abstraction for the offered products and a standardized way to express all of the application requirements for the resulting deployments. In this paper, we contribute by making yet another case for multi-cloud deployments and by predicting the emergence of a new generation of application-level resource managers which will natively support multi-cloud for enterprise applications. We identify three main components of the feedback loop controlled application-level resource managers: the software life-cycle manager, the data storage and access manager, and the service execution manager.

[1]  Eui-nam Huh,et al.  Efficient Service Recommendation System for Cloud Computing Market , 2009, FGIT-GDC.

[2]  Michael Hamburg,et al.  Meltdown: Reading Kernel Memory from User Space , 2018, USENIX Security Symposium.

[3]  Andy B. Yoo,et al.  Approved for Public Release; Further Dissemination Unlimited X-ray Pulse Compression Using Strained Crystals X-ray Pulse Compression Using Strained Crystals , 2002 .

[4]  Dharmesh Kakadia,et al.  Apache Mesos Essentials , 2015 .

[5]  Lúcia Maria de A. Drummond,et al.  Optimization of a Cloud Resource Management Problem from a Consumer Perspective , 2013, Euro-Par Workshops.

[6]  Sanjay Ghemawat,et al.  MapReduce: Simplified Data Processing on Large Clusters , 2004, OSDI.

[7]  Randy H. Katz,et al.  Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center , 2011, NSDI.

[8]  Carlo Curino,et al.  Apache Hadoop YARN: yet another resource negotiator , 2013, SoCC.

[9]  Romain Rouvoy,et al.  A Federated Multi-cloud PaaS Infrastructure , 2012, 2012 IEEE Fifth International Conference on Cloud Computing.

[10]  Randy H. Katz,et al.  The Datacenter Needs an Operating System , 2011, HotCloud.

[11]  David A. Patterson,et al.  In-datacenter performance analysis of a tensor processing unit , 2017, 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA).

[12]  Abhishek Verma,et al.  Large-scale cluster management at Google with Borg , 2015, EuroSys.

[13]  Pete Wyckoff,et al.  Hive - A Warehousing Solution Over a Map-Reduce Framework , 2009, Proc. VLDB Endow..

[14]  Marin Litoiu,et al.  Introducing STRATOS: A Cloud Broker Service , 2012, 2012 IEEE Fifth International Conference on Cloud Computing.

[15]  Jane Siegel,et al.  Cloud Services Measures for Global Use: The Service Measurement Index (SMI) , 2012, 2012 Annual SRII Global Conference.

[16]  Scott Shenker,et al.  Spark: Cluster Computing with Working Sets , 2010, HotCloud.

[17]  Dana Petcu,et al.  Multi-Cloud: expectations and current approaches , 2013, MultiCloud '13.

[18]  Rajkumar Buyya,et al.  High Performance Cluster Computing: Architectures and Systems , 1999 .

[19]  Garrick Staples,et al.  TORQUE resource manager , 2006, SC.

[20]  Hairong Kuang,et al.  The Hadoop Distributed File System , 2010, 2010 IEEE 26th Symposium on Mass Storage Systems and Technologies (MSST).

[21]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[22]  Howard Gobioff,et al.  The Google file system , 2003, SOSP '03.