The ongoing adoption of Cloud Computing at a fast rate has lead to an increase in the number of users and in the same time, in the level of complexity and performance. The interaction model is based on services offered to users at different levels. Metal-as-a-service (MaaS) platforms assure a higher level of performance when compared to typical Cloud platforms, but at the cost of a more complex provisioning system. An advanced reservation and provisioning system for MaaS should alleviate the problem of added complexity by accounting for the observed deployment time of the infrastructures. The paper studies the provisioning capabilities of the Bigstep Full Metal Cloud platform in order to allow the construction of such a system. The main issue addressed is oriented on cloud management automation, SLA and QoS management.
{"title":"MaaS advanced provisioning and reservation system","authors":"Alexandru Sirbu, C. Pop, Florin Pop","doi":"10.1145/2747470.2747473","DOIUrl":"https://doi.org/10.1145/2747470.2747473","url":null,"abstract":"The ongoing adoption of Cloud Computing at a fast rate has lead to an increase in the number of users and in the same time, in the level of complexity and performance. The interaction model is based on services offered to users at different levels. Metal-as-a-service (MaaS) platforms assure a higher level of performance when compared to typical Cloud platforms, but at the cost of a more complex provisioning system. An advanced reservation and provisioning system for MaaS should alleviate the problem of added complexity by accounting for the observed deployment time of the infrastructures. The paper studies the provisioning capabilities of the Bigstep Full Metal Cloud platform in order to allow the construction of such a system. The main issue addressed is oriented on cloud management automation, SLA and QoS management.","PeriodicalId":328734,"journal":{"name":"AIMC '15","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128301313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As cloud-deployed applications became more and more mainstream, continuously more complex services started to be deployed; indeed where initially monolithic applications were simply ported to the cloud, applications are now more and more often composed of micro-services. This improves the flexibility of an application but also makes it more complex due to the sheer number of services comprising it. As deployment and runtime management becomes more complex, orchestration software are becoming necessary to completely manage the lifecycle of cloud applications. One crucial problem remaining is how these applications can be made reliable in the cloud, a naturally unreliable environment. In this paper we propose concepts and architectures which were implemented in our orchestration software to guarantee reliability. Our initial implementation also relies on Monasca, a well-known monitoring software for Open-Stack, to gather proper metric and execute threshold-based actions. This allows us to show how service reliability can be ensured using orchestration and how a proper incident-management software feeding decisions to the orchestration engine ensures high-availability of all components of managed applications.
{"title":"Reliable cloud-applications: an implementation through service orchestration","authors":"Florian Dudouet, A. Edmonds, Michael Erne","doi":"10.1145/2747470.2747471","DOIUrl":"https://doi.org/10.1145/2747470.2747471","url":null,"abstract":"As cloud-deployed applications became more and more mainstream, continuously more complex services started to be deployed; indeed where initially monolithic applications were simply ported to the cloud, applications are now more and more often composed of micro-services. This improves the flexibility of an application but also makes it more complex due to the sheer number of services comprising it.\u0000 As deployment and runtime management becomes more complex, orchestration software are becoming necessary to completely manage the lifecycle of cloud applications. One crucial problem remaining is how these applications can be made reliable in the cloud, a naturally unreliable environment.\u0000 In this paper we propose concepts and architectures which were implemented in our orchestration software to guarantee reliability. Our initial implementation also relies on Monasca, a well-known monitoring software for Open-Stack, to gather proper metric and execute threshold-based actions. This allows us to show how service reliability can be ensured using orchestration and how a proper incident-management software feeding decisions to the orchestration engine ensures high-availability of all components of managed applications.","PeriodicalId":328734,"journal":{"name":"AIMC '15","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132214385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
G. T. Carughi, Sandro Brunner, Martin Blöchlinger, Florian Dudouet, A. Edmonds
Running applications in the cloud efficiently requires much more than deploying software in virtual machines. Cloud applications have to be continuously managed: 1) to adjust their resources to the incoming load and 2) to face transient failures replicating and restarting components to provide resiliency on unreliable infrastructure. Continuous management monitors application and infrastructural metrics to provide automated and responsive reactions to failures (health management) and changing environmental conditions (auto-scaling) minimizing human intervention. In the current practice, management functionalities are provided as infrastructural or third party services. In both cases they are external to the application deployment. We claim that this approach has intrinsic limits, namely that separating management functionalities from the application prevents them from naturally scaling with the application and requires additional management code and human intervention. Moreover, using infrastructure provider services for management functionalities results in vendor lock-in effectively preventing cloud applications to adapt and run on the most effective cloud for the job. In this position paper we propose a novel architecture that enables scalable and resilient self-management of microservices applications on cloud.
{"title":"An architecture for self-managing microservices","authors":"G. T. Carughi, Sandro Brunner, Martin Blöchlinger, Florian Dudouet, A. Edmonds","doi":"10.1145/2747470.2747474","DOIUrl":"https://doi.org/10.1145/2747470.2747474","url":null,"abstract":"Running applications in the cloud efficiently requires much more than deploying software in virtual machines. Cloud applications have to be continuously managed: 1) to adjust their resources to the incoming load and 2) to face transient failures replicating and restarting components to provide resiliency on unreliable infrastructure. Continuous management monitors application and infrastructural metrics to provide automated and responsive reactions to failures (health management) and changing environmental conditions (auto-scaling) minimizing human intervention.\u0000 In the current practice, management functionalities are provided as infrastructural or third party services. In both cases they are external to the application deployment. We claim that this approach has intrinsic limits, namely that separating management functionalities from the application prevents them from naturally scaling with the application and requires additional management code and human intervention. Moreover, using infrastructure provider services for management functionalities results in vendor lock-in effectively preventing cloud applications to adapt and run on the most effective cloud for the job.\u0000 In this position paper we propose a novel architecture that enables scalable and resilient self-management of microservices applications on cloud.","PeriodicalId":328734,"journal":{"name":"AIMC '15","volume":"14 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113932757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we present a scalable parallel virtual machine planning and failover method for high availability at a virtual machine (VM) level in a data center. The solution is implemented in IBM's Cloud Managed Services (CMS) enterprise cloud offering for rapid failover in large data centers with a large number of servers, VMs, and disks. The failover system enables failover-time planning and execution and keeps the recovery time within limits of service level agreement (SLA) allowed time budget. The initial serial failover time is reduced by a factor of up to 11 for parallel implementation, and by a factor of up to 44 for parallel failover - parallel storage mapping implementation.
{"title":"ResilientVM: high performance virtual machine recovery in the cloud","authors":"V. Salapura, R. Harper","doi":"10.1145/2747470.2747472","DOIUrl":"https://doi.org/10.1145/2747470.2747472","url":null,"abstract":"In this paper, we present a scalable parallel virtual machine planning and failover method for high availability at a virtual machine (VM) level in a data center. The solution is implemented in IBM's Cloud Managed Services (CMS) enterprise cloud offering for rapid failover in large data centers with a large number of servers, VMs, and disks. The failover system enables failover-time planning and execution and keeps the recovery time within limits of service level agreement (SLA) allowed time budget. The initial serial failover time is reduced by a factor of up to 11 for parallel implementation, and by a factor of up to 44 for parallel failover - parallel storage mapping implementation.","PeriodicalId":328734,"journal":{"name":"AIMC '15","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124929900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：