Data Center Consolidation Research Articles

MultiLanes

OS-level virtualization is often used for server consolidation in data centers because of its high efficiency. However, the sharing of storage stack services among the colocated containers incurs contention on shared kernel data structures and locks within I/O stack, leading to severe performance degradation on manycore platforms incorporating fast storage technologies (e.g., SSDs based on nonvolatile memories). This article presents MultiLanes, a virtualized storage system for OS-level virtualization on manycores. MultiLanes builds an isolated I/O stack on top of a virtualized storage device for each container to eliminate contention on kernel data structures and locks between them, thus scaling them to manycores. Meanwhile, we propose a set of techniques to tune the overhead induced by storage-device virtualization to be negligible, and to scale the virtualized devices to manycores on the host, which itself scales poorly. To reduce the contention within each single container, we further propose SFS, which runs multiple file-system instances through the proposed virtualized storage devices, distributes all files under each directory among the underlying file-system instances, then stacks a unified namespace on top of them. The evaluation of our prototype system built for Linux container (LXC) on a 32-core machine with both a RAM disk and a modern flash-based SSD demonstrates that MultiLanes scales much better than Linux in micro- and macro-benchmarks, bringing significant performance improvements, and that MultiLanes with SFS can further reduce the contention within each single container.

Data Center Consolidation at the University at Albany

This paper describes the experience of the University at Albany (UAlbany) Libraries’ migration to a centralized University data center. Following an introduction to the environment at UAlbany, the authors discuss the advantages of data center consolidation. Lessons learned from the project include the need to participate in the planning process, review migration schedules carefully, clarify costs of centralization, agree on a service level agreement, communicate plans to customers, and leverage economies of scale.

A New Approach for Dynamic Virtual Machine Consolidation in Cloud Data Centers

Cloud computing environments have introduced a new model of computing by shifting the location of computing infrastructure to the Internet network to reduce the cost associated with the management of hardware and software resources. The Cloud model uses virtualization technology to effectively consolidate virtual machines (VMs) into physical machines (PMs) to improve the utilization of PMs. Studies however have shown that the average utilization of PMs in many Cloud data centers is still lower than expected. The Cloud model is expected to improve the existing level of utilization by employing new approaches of consolidation mechanisms. In this paper we propose a new approach for dynamic consolidation of VMs in order to maximize the utilization of PMs. This is achieved by a dynamic programing algorithm that selects the best VMs for migration from an overloaded PM, considering the migration overhead of a VM. Evaluation results demonstrate that our algorithms achieve good performance.

Cost-Benefit Analysis of Datacenter Consolidation Using Virtualization

The objectives of datacenter consolidation are cost reduction and superior services. A datacenter consolidation plan includes minimizing investments in IT infrastructure and buildings and reducing power consumption related to cooling. Such a process requires scalable planning and implementation. Virtualization is the most popular and cost-effective technology for datacenter consolidation. In this article, the author runs a cost-benefit analysis of virtualization and datacenter consolidation using the Global Virtual datacenter online calculator and VMware's ROI CO ((for) return on investment/total cost of ownership) calculator version 3.0.

Efficient memory virtualization for Cross-ISA system mode emulation

Cross-ISA system-mode emulation has many important applications. For example, Cross-ISA system-mode emulation helps computer architects and OS developers trace and debug kernel execution-flow efficiently by emulating a slower platform (such as ARM) on a more powerful plat-form (such as an x86 machine). Cross-ISA system-mode emulation also enables workload consolidation in data centers with platforms of different instruction-set architectures (ISAs). However, system-mode emulation is much slower. One major overhead in system-mode emulation is the multi-level memory address translation that maps guest virtual address to host physical address. Shadow page tables (SPT) have been used to reduce such overheads, but primarily for same-ISA virtualization. In this paper we propose a novel approach called embedded shadow page tables (ESPT). EPST embeds a shadow page table into the address space of a cross-ISA dynamic binary translation (DBT) and uses hardware memory management unit in the CPU to translate memory addresses, instead of software translation in a current DBT emulator like QEMU. We also use the larger address space on modern 64-bit CPUs to accommodate our DBT emulator so that it will not interfere with the guest operating system. We incorporate our new scheme into QEMU, a popular, retargetable cross-ISA system emulator. SPEC CINT2006 benchmark results indicate that our technique achieves an average speedup of 1.51 times in system mode when emulating ARM on x86, and a 1.59 times speedup for emulating IA32 on x86_64.

Transforming the information technology infrastructure of IBM

The structure, strategy, and business operation of IBM have experienced a significant transformation over the last several years. This transformation has been accompanied by an equally important transformation of the internal services and technologies that support the company. Key among such activities has been the renovation of company information technology (IT) and the infrastructure it is built on. This paper describes the transformation of the internal IT infrastructure of IBM and some of the key initiatives that contributed to a smarter, optimized capability. We describe the activities, best practices, and lessons learned in server virtualization work, cloud computing incubation and evolution, workload location and data-center consolidation, and the transformation of the internal communications network of IBM. Realizing that transformation is an on-going process, we conclude the paper with an exploration of our plans to use analytics to enable the expansion and delivery of a real-time collaboration experience, another key IT service area.

Modeling response times in the Google ROADEF/EURO challenge

In this paper, we extend the machine reassignment model proposed by Google for the ROADEF/EURO Challenge. The aim of the challenge is to develop algorithms for the efficient solutions of data-center consolidation problems. The problem stated in the challenge mainly focus on dependability requirements and does not take into account performance requirements (end-to-end response times). We extend the Google problem definition by modeling and constraining end-to-end response times. We provide experimental results to show the effectiveness of this extension.

CloudBridge

Data center consolidation has emerged as an important tool to improve the hardware utilization of data centers and reduce delivery costs. Consolidation has traditionally used virtualization to consolidate multiple workloads as different virtual machines running on a shared physical server. Consolidation leads to reduction in the hardware and the facilities cost (e.g., space and energy) but does not reduce software maintenance cost, which is often proportional to the number of instances of the software. The ever-increasing proportion of software cost (labour and license) in data center operations make software consolidation an important tool in data centers. In this work, we investigate the problem of data center consolidation with the goal of minimizing the total costs in running a data center. We build on existing work on virtualization-driven hardware consolidation and software consolidation to design CloudBridge that reduces total data center cost. CloudBridge uses an algorithm for finding the optimal software consolidation level for workloads consolidated on a physical server. Further, it intelligently optimizes the tradeoff between hardware and software consolidation to identify suitable workloads that should be co-located on a shared physical server. We present both theoretical and experimental evidence that establishes the effectiveness of CloudBridge.

A Strategic Approach to Network Defense: Framing the Cloud

With cheap technology and minimal investment, current and potential adversaries operating in cyberspace can inflict serious damage to DOD's vast information grid system that encompasses more than 15,000 local, regional, and wide-area networks, and approximately 7 million information technology devices. (1) --Robert Gates, former US Secretary of Defense The US Government has robust data networks that provide rapid transport of imagery, textual information, command and control data, and routine communications to support military operations and core business needs. This information is vital in the conduct of its war and peacetime missions. Historically, America's adversaries attempt to leverage network vulnerabilities to gain strategic advantage by exploiting information about US military and commercial activities, trade secrets, financial information, system architectures, and other data. The US is arguably the most interconnected nation on earth and it plays hegemonic role with regard to establishing and maintaining the rules that govern the Internet. Americans embrace digital technologies that promise greater interconnection for governmental, corporate, and personal utility. This article examines current Internet attack trends in the computer networking environment and proposes an enhanced framework for strategic system defense applicable to both corporate and federal networks. Presently, the balance of power favors those adversaries trying to attack US information systems, networks, and critical infrastructure. Well-designed cloud computing environments, however, may change the balance in favor of the defense, while reducing costs and improving service. The enhanced framework addresses these issues and assists in reducing the risks associated with assessing and adopting cloud computing. Computing clouds are large data centers, filled with generic processing and storage facilities, and operated as multiple reconfigurable virtual servers. (2) Traditionally, cloud computing was represented by the outsourcing of an organization's computing infrastructure. Today, cloud computing presents a model for enabling ubiquitous, convenient, on-demand network access to shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. (3) Why Cloud Computing? Emerging cloud computing technologies will subsume existing enterprise networks and encompass system defenses that are designed, implemented, and managed at corporate information technology (IT) processing centers. Once applications are logically extended through virtualization in cloud computing environment, they are no longer tied to physical location. The cloud service provider develops dispersed support and hosting facilities that allow applications to perform as needed. The system user merely needs to access the typically web-based application to run the desired program. The trend for networking infrastructures and computing centers is shifting toward consolidation for cost savings. Cloud computing provides for the outsourcing &entire data centers, the saving of physical space, infrastructure, and labor costs. The prime benefit is the reduced cost of updating information systems and infrastructures, which is transferred to the cloud provider. (4) Cloud computing is major evolutionary leap that virtualizes servers, infrastructures, and software as pay-for-use services. Government leaders have identified the benefits gained by adopting cloud computing, but they have not adequately considered the inherent risk with outsourcing IT. Envisioning the future, the US Chief Information Officer (CIO) announced the Federal Data Center Consolidation Initiative and issued instructions for the Federal CIO Council to have governmental departments inventory their IT assets and integrate consolidation plans into their 2012 planning budgets. …

Successful Introduction of Cloud Computing into your Organization: A Six-Step Conceptual Model

LITERATURE SURVEY: THE LANDSCAPE OF CLOUD COMPUTING A recent survey on the status, applications, versatilities, popularities, and opportunities gained by using cloud computing platforms reveals interesting facts as follows: Many banks including ING are testing cloud computing for integrating it into their operations. In a survey of 186 banking technology professionals interested in cloud computing 73 percent of the participants indicated that cloud computing will enable them to meet user demands quickly and achieve scale (Crosman, 2010). Morgan Stanley is another major bank that is planning to integrate cloud computing into its operations. However, security and privacy in cloud computing are among the major concerns (Cohen, 2008). Department of Defense (DoD) is exploring the use of cloud computing in one of the following four categories (Anonymous, 2010): * Use commercially provided cloud services * Deploy cloud computing within DoD networks * Develop a multi-agency cloud computing network for data processing and storage * Develop a combined DoD/commercial provider system Key decision makers believe cloud computing will provide greater agility and flexibility for the DoD, today and in the future. Cloud computing will also increase efficiencies and incur marked cost savings during their life-cycles to alleviate some of the pressure of budget reductions. Once again, security and privacy are major concerns for this deployment (Forbes, 2011, Anonymous, 2011). In addition, experts believe that, the department of defense could run more effectively using cloud technologies. DoD could become a model and show other large organizations how cloud computing applications could be used (Linthicum, 2010d). According to experts it makes no sense for smaller enterprises to operate their own data centers. Cloud computing will be both cost effective and also environmentally safer and cleaner for such organizations (Krill, 2010). Cloud computing platforms clearly support the growing phenomenal of computing and could help organization to achieve their green goals. An executive at global consultancy CSC predicts that half of all IT workloads will become cloud services by 2016 (Knorr, 2011). Despite millisecond delays cross-ocean cloud applications are gaining in popularity because of major cost saving and other advantages (Thibodeau, 2011c). President Barack Obama relies heavily on cloud computing and data center consolidation to keep IT costs under control (Thibodeau, 2011b). The Obama administration aims to cut costs by embracing the cloud in place of in-house, proprietary technology (Samson, 2010). A report published in Techland.time.com in February 2011 indicates that YouTube uses cloud computing infrastructure to split video processing between hundreds of thousands of machines, allowing them to process video seven times faster than in 2008 (and even four times faster than just six months ago) (McMillan, 2011). Tata Communications will offer cloud computing that includes services for computing, data transfer, and storage (Ribeiro, 2010). Joining Microsoft, Oracle, Google and others, IBM plans to offer a cloud-based office suite (Jackson, 2011). IBM also puts Notes and collaboration in the cloud (Knorr, 2010). China is building a city-sized cloud computing and office complex that will include a mega data center. Cloud computing is among the projects supporting that country's double-digit growth in IT spending (Thibodeau, 2011a). BACKGROUND As reviewed in the previous section, cloud computing is one of the fastest area of application within the information systems field. Most computer and networking companies are trying to either enter into this area or increase their existing market share. Even during the economic down turn, cloud computing related jobs have been in high demand (Linthicum, 2010b). …

JM Family Enterprises Inc.: Selectively Outsourcing IT for Increased Business Value

In 2001, JM Family Enterprises (JMFE), the 15th largest privately held company in the United States, began a transformation of its IT services. By 2004, the company was ranked seventh in InformationWeek magazine's listing of most innovative users of IT. JMFE's IT transformation involved rationalization of its technology architecture, adoption of best service practices, data center consolidation and selective outsourcing. This case study emphasizes the role of outsourcing in JMFE's IT transformation. Executives from both JMFE and IBM describe how they have managed the relationship to ensure mutual benefits.

The Los Angeles County Experience

Abstract The author provides a recounting and analysis of his experience for the past decade as Chairman of the Information Systems Commission of the Los Angeles County Board of Supervisors. The Commission is an advisory body to the Board of Supervisors regarding information technology for which the County spends more than 700 million dollars annually. Issues covered by the Commission have included creation of a CIO function, Y2K activity, large-scale systems development projects, ERP, IT position reclassification and data center consolidation. These are representative of information technology issues faced by local governments in the United States.