Waheed Iqbal

Adaptive resource provisioning for read intensive multi-tier applications in the cloud

A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers of services. However, current SLAs offered by cloud infrastructure providers do not address response time, which, from the users point of view, is the most important quality attribute for Web applications. Satisfying a maximum average response time guarantee for Web applications is difficult for two main reasons: first, traffic patterns are highly dynamic and difficult to predict accurately; second, the complex nature of multi-tier Web applications increases the difficulty of identifying bottlenecks and resolving them automatically. This paper proposes a methodology and presents a working prototype system for automatic detection and resolution of bottlenecks in a multi-tier Web application hosted on a cloud in order to satisfy specific maximum response time requirements. It also proposes a method for identifying and retracting over-provisioned resources in multi-tier cloud-hosted Web applications. We demonstrate the feasibility of the approach in an experimental evaluation with a testbed EUCALYPTUS-based cloud and a synthetic workload. Automatic bottleneck detection and resolution under dynamic resource management has the potential to enable cloud infrastructure providers to provide SLAs for Web applications that guarantee specific response time requirements while minimizing resource utilization.

SLA-Driven Adaptive Resource Management for Web Applications on a Heterogeneous Compute Cloud

Current service-level agreements (SLAs) offered by cloud providers make guarantees about quality attributes such as availability. However, although one of the most important quality attributes from the perspective of the users of a cloud-based Web application is its response time, current SLAs do not guarantee response time. Satisfying a maximum average response time guarantee for Web applications is difficult due to unpredictable traffic patterns, but in this paper we show how it can be accomplished through dynamic resource allocation in a virtual Web farm. We present the design and implementation of a working prototype built on a EUCALYPTUS-based heterogeneous compute cloud that actively monitors the response time of each virtual machine assigned to the farm and adaptively scales up the application to satisfy a SLA promising a specific average response time. We demonstrate the feasibility of the approach in an experimental evaluation with a testbed cloud and a synthetic workload. Adaptive resource management has the potential to increase the usability of Web applications while maximizing resource utilization.

SLA-Driven Dynamic Resource Management for Multi-tier Web Applications in a Cloud

Current service-level agreements (SLAs) offered by cloud providers do not make guarantees about response time of Web applications hosted on the cloud. Satisfying a maximum average response time guarantee for Web applications is difficult due to unpredictable traffic patterns. The complex nature of multi-tier Web applications increases the difficulty of identifying bottlenecks and resolving them automatically. It may be possible to minimize the probability that tiers (hosted on virtual machines) become bottlenecks by optimizing the placement of the virtual machines in a cloud. This research focuses on enabling clouds to offer multi-tier Web application owners maximum response time guarantees while minimizing resource utilization. We present our basic approach, preliminary experiments, and results on a EUCALYPTUS-based testbed cloud. Our preliminary results shows that dynamic bottleneck detection and resolution for multi-tier Web application hosted on the cloud will help to offer SLAs that can offer response time guarantees.

SLA-Driven automatic bottleneck detection and resolution for read intensive multi-tier applications hosted on a cloud

A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers of services However, the current SLAs offered by cloud providers do not address response time, which, from the users point of view, is the most important quality attribute for Web applications Satisfying a maximum average response time guarantee for Web applications is difficult for two main reasons: first, traffic patterns are unpredictable; second, the complex nature of multi-tier Web applications increases the difficulty of identifying bottlenecks and resolving them automatically This paper presents a working prototype system that automatically detects and resolves bottlenecks in a multi-tier Web application hosted on a EUCALYPTUS-based cloud in order to satisfy specific maximum response time requirements We demonstrate the feasibility of the approach in an experimental evaluation with a testbed cloud and a synthetic workload Automatic bottleneck detection and resolution under dynamic resource management has the potential to enable cloud providers to provide SLAs for Web applications that guarantee specific response time requirements.

Black-box approach to capacity identification for multi-tier applications hosted on virtualized platforms

In cloud-based Web application hosting environments, virtualization offers the potential to exploit dynamic resource provisioning and scaling to maintain service level agreements while minimizing resource utilization for a given workload. However, optimal proactive resource provisioning and scaling for a specific Web application require, at the least, a profile of the applications current workload and a model of the applications capacity under various resource configurations. Here we focus on multi-tier Web applications. The capacity of a multi-tier Web application varies substantially as the pattern of requests in the workload changes. In this paper, we propose and evaluate a black-box method for capacity prediction that first identifies workload patterns for a multi-tier Web application from access logs using unsupervised machine learning and then, based on those patterns, builds a model capable of predicting the applications capacity for any specific workload pattern. In an experimental evaluation, we compare a baseline method that predicts capacity without a model of the application-specific workload patterns to several regression models using the proposed workload identification method. All of the models based on workload pattern identification outperform the baseline method. The best model, a Gaussian process regression model, gives only 6.42% error. Cloud providers utilizing our method can proactively perform dynamic allocation of resources to multi-tier Web applications, meeting service level agreements at minimal cost.

Low Cost Quality Aware Multi-tier Application Hosting on the Amazon Cloud

Public cloud providers offer a variety of services and resources enabling users to host their applications. In order to determine the set of resources that will best minimize hosting costs while simultaneously meeting performance requirements, cloud users must be aware of the cost-performance tradeoffs of the available resources. This is particularly true of multi-tier Web applications using dynamic scaling strategies to sustain specific response time requirements. In this paper, we provide a cost-performance analysis of multiple scale out strategies using the three most economically feasible Amazon EC2 compute resources (micro, small, and medium virtual machine instances) for a typical multi-tier Web application. We find that the special CPU allocation policy of the micro instance makes it especially suitable for satisfying Web application performance requirements at minimal cost on the Amazon cloud.

On the use of CryptDB for securing Electronic Health data in the cloud: A performance study

Electronic Health Records (EHRs) are the very personal data and ensuring its privacy and security is an utmost priority. Various laws require the privacy of this data to be ensured and usually this is achieved using strict access control methods. However, these methods have limitations, specifically in case of a server breach. In this paper, we use CryptDB to ensure data confidentiality in EHR Systems. In particular, we investigate the performance of CryptDB with OpenEMR, on different deployment scenarios and varying workloads over the cloud and a local testbed. We identify that CryptDB successfully provides the data confidentiality on the database server when deployed on the cloud. We also find that for a mix workload, the average performance of the OpenEMR with CryptDB in the cloud remains under two seconds which makes CryptDB a viable option for providing security to EHR systems deployed in the cloud. This is the first study to integrate CryptDB with OpenEMR and to profile performance overhead to ensure the data confidentiality under different deployment and varying workload scenarios in the cloud.

Minimalistic Adaptive Resource Management for Multi-tier Applications Hosted on Clouds

Cloud computing allows Web application owners to host their applications with low operational cost and enables them to scale applications to maintain performance based on traffic load and application resource requirements. However, for multi-tier Web applications, it is difficult to automatically identify the exact location of a bottleneck and scale the appropriate resource tier accordingly because multi-tier applications are complex and bottleneck patterns may be dependent on the specific pattern of workload at any given time. This Ph.D. dissertation aims to explore the possibilities to satisfy response time guarantees for multi-tier applications hosted on clouds using adaptive resource management with minimal hardware profiling and application-centric knowledge.

AINAN: Collaborative Application for Protein Visualization & Analyses

AINAAN is a collaborative application developed for molecular visualization and analysis with multiple view points. The application allows multiple users to view and manipulate single and multiple representations of molecular structures. Network sessions can be initiated by any user over the network. Members over the network can join the session, and the initiator owns the activities of the session. The application provides conventional molecular visualization facilities like rendering, rotation, with different views, along with the usage of protein network stores. Protein store is the repository of the protein structures, alignment results, and movies maintained by each user, which can be shared over the network. Sequence and structural analysis algorithms are also supported in AINAAN. Sessions can be saved by recording events and alignment results for future reference. Large scale study of protein molecular structures will need network views support for comparative analysis.

On Providing Response Time Guarantees to a Cloud-Hosted Telemedicine Web Service

Traditionally healthcare services are deployed on dedicated physical systems and the functionalities are limited to the local network. Mostly, dedicated physical systems are either under-provisioned or over-provisioned. Cloud Computing technology addresses these limitations by dynamically allocating required resources to applications being hosted on such cloud platforms. In this paper, we study the viability of hosting a telemedicine service over Amazon Elastic Compute Cloud (EC2); a public cloud architecture. In particular, we study the performance of our telemedicine service under linearly increasing workloads by using multiple hosting options available in Amazon EC2. The performance analysis of our telemedicine service is based on fulfilling the specific number of requests per seconds under constraint response times. We find that dynamic resource provisioning on the web tier using medium type instances gives better results compared to static allocation using large and xlarge type instances without incurring any bottleneck issues, thereby, making it a feasible solution for telemedicine service providers.