Data Center Parameters Research Articles

Modeling the Performance of the Clock Phase Caching Approach to Clock and Data Recovery

Optical switching could enable data center networks to keep pace with the rapid growth of intra-data center traffic, however, sub-nanosecond clock and data recovery time is crucial to enabling optically-switched data center networks to transport small packet dominated data center traffic with over 90% efficiency. We review the clock-synchronized approach to clock and data recovery, which enables sub-nanosecond switching time in optically switched networks. We then introduce an analytical model to mathematically explore the operation of clock phase caching, and use this model to explore the impact of factors such as fiber temperature, clock jitter and symbol rate on the BER and clock and data recovery locking time performance of the clock phase caching approach, as well as their impact on scalability. Using commercial data center parameters matching those used in our previous experimental research, we find that our analytical model provides estimates that closely match our previous experimental results, validating its use for making predictions of the performance of clock phase cached systems.

Simulation-Based Assessment of Performance Indicator for Data Center Cooling Optimization

Abstract Data center power consumption has grown substantially in the past 20 years. According to the United States Data Center Energy Usage Report, the data center consumption in 2014 was estimated at 70 billion kWh, which accounted for 1.8% of the total U.S. electricity. The present effort investigates the effects of various data center parameters on a new set of metrics called the performance indicator to help assess and optimize the cooling performance of data centers. The three metrics in the performance indicator include power usage effectiveness ratio (PUEr), thermal conformance, and thermal resilience. The data center parameters investigated include computer room air handler (CRAH) setpoints, room configuration layouts, and containment strategies. The results show that the CRAH setpoint significantly influences the PUEr with higher setpoint values resulting in lower PUEr values. Room configuration layout changes and containment strategies showed substantial effects on thermal conformance and thermal resilience. The thermal conformance was increased approximately 10% with room configuration changes without changing the PUEr. Full hot aisle containment also improved the thermal conformance by 7.4%.

Study of optimization parameters for service chaining in cloud environment

and it is growing towards integration of 5G technology in near future. Therefore, to improve the quality of experience for viewing the content over the internet requires dynamic allocation and adaptation of network resources in an optimized manner. Traditional IP networks are vertically integrated hence flexibility in network resources management is very less. Software-defined networking (SDN) as an emerging technology, which comes with the promise of the solution to dynamically govern various network resources by breaking this chain or hierarchy of vertical integration. Network function virtualization along with service chain optimization provides the solution to enhance the Quality of Experience (QoE) and Quality of Service (QoS). In this paper, we are proposing an approach to improve the QoE by ameliorating the service chain and data center parameters.

Performance Evaluation of Cloud Computing Resources

Cloud computing is an emerging information technology which is rapidly growing. However, measuring the performance of cloud based applications in real environments is a challenging task for research as well as business community. In this work, we focused on Infrastructure as a Service (IaaS) facility of cloud computing. We made a performance evaluation of two renowned public and private cloud platforms. Several performance metrics such as integer, floating Point, GFLOPS, read, random Read, write, random write, bandwidth, jitter and throughput were used to analyze the performance of cloud resources. The motive of this analysis is to help cloud providers to adjust their data center parameters under different working conditions as well as cloud customers to monitor their hired resources. We analyzed and compared the performance of OpenStack and Windows Azure platforms by considering resources like CPU, memory, disk and network in a real cloud setup. In order to evaluate each feature, we used related benchmarks, for example, Geekbench & LINPACK for CPU performance, RAMspeed & STREAM for memory performance, IOzone for disk performance and Iperf for network performance. Our experimental results showed that the performance of both clouds is almost same; however, OpenStack seems to be better option as compared to Windows Azur keeping in view its cost as well as network performance.

ASIC Clouds: Specializing the Datacenter

GPU and FPGA-based clouds have already demonstrated the promise of accelerating computing-intensive workloads with greatly improved power and performance.In this paper, we examine the design of ASIC Clouds, which are purpose-built datacenters comprised of large arrays of ASIC accelerators, whose purpose is to optimize the total cost of ownership (TCO) of large, high-volume chronic computations, which are becoming increasingly common as more and more services are built around the Cloud model. On the surface, the creation of ASIC clouds may seem highly improbable due to high NREs and the inflexibility of ASICs. Surprisingly, however, large-scale ASIC Clouds have already been deployed widely to implement the distributed Bitcoin cryptocurrency system.Among our contributions, we present a methodology that given an accelerator design, derives Pareto-optimal ASIC cloud Servers, by extracting data from place-and-routed circuits and computational fluid dynamic simulations, and then employing clever but brute-force search to find the best jointly-optimized ASIC, DRAM subsystem, motherboard, power delivery system, cooling system, operating voltage, and case design. Moreover, we show how data center parameters determine which of the many Pareto-optimal points is TCO-optimal. Finally we examine when it makes sense to build an ASIC Cloud, and examine the impact of ASIC NRE.

ASIC clouds

GPU and FPGA-based clouds have already demonstrated the promise of accelerating computing-intensive workloads with greatly improved power and performance. In this paper, we examine the design of ASIC Clouds, which are purpose-built datacenters comprised of large arrays of ASIC accelerators, whose purpose is to optimize the total cost of ownership (TCO) of large, high-volume chronic computations, which are becoming increasingly common as more and more services are built around the Cloud model. On the surface, the creation of ASIC clouds may seem highly improbable due to high NREs and the inflexibility of ASICs. Surprisingly, however, large-scale ASIC Clouds have already been deployed by a large number of commercial entities, to implement the distributed Bitcoin cryptocurrency system. We begin with a case study of Bitcoin mining ASIC Clouds, which are perhaps the largest ASIC Clouds to date. From there, we design three more ASIC Clouds, including a YouTube-style video transcoding ASIC Cloud, a Litecoin ASIC Cloud, and a Convolutional Neural Network ASIC Cloud and show 2-3 orders of magnitude better TCO versus CPU and GPU. Among our contributions, we present a methodology that given an accelerator design, derives Pareto-optimal ASIC Cloud Servers, by extracting data from place-and-routed circuits and computational fluid dynamic simulations, and then employing clever but brute-force search to find the best jointly-optimized ASIC, DRAM subsystem, motherboard, power delivery system, cooling system, operating voltage, and case design. Moreover, we show how data center parameters determine which of the many Pareto-optimal points is TCO-optimal. Finally we examine when it makes sense to build an ASIC Cloud, and examine the impact of ASIC NRE.

The Greening of Data Centers with Cloud Technology

This article presents a decision support system to provide green or energy efficient solutions for data centers that maintain computers and peripherals to serve organizations. Traditionally, data centers catered to all operations using in-house servers. Cloud technology provides alternatives to outsource operations heading towards greenness. However, using cloud services for all data center operations may have its pitfalls. In this paper, the authors analyze various data center parameters such as carbon footprint and power usage effectiveness along with cloud-based and server-based models. They consider data mining techniques of decision trees and case based reasoning in their work. Among other findings, they head towards a hybrid model that meets the demands of productivity, energy efficiency and related factors. These findings lead to the development of the decision support system. The authors describe the research, development and evaluation of the system. They conclude with important outcomes deployed in real-world scenarios in data center management.

English

Cloud Computing provides scalable computing and storage resources. More and more data-intensive applications are developed in this computing environment.Evaluation of cloud data center performance is a key attribute in propagation of IaaS cloud services. This evaluation is necessary to quantify the cost benefit and Quality Of Service (QoS) of Cloud services. Cloud system managers find it difficult to maintain performance data due to the huge number of parameters that need to be monitored in a cloud system. Different applications have different quality-ofservice (QoS) requirements. In this paper, we present an analytical model, based on Stochastic Reward Nets (SRNs), that is both scalable to model systems composed of thousands of resources and flexible to represent different policies and cloud-specific strategies. Several performance metrics are defined and evaluated to analyze the behavior of a Cloud data center: utilization, availability, waiting time, and responsiveness. We are also notifying the resiliency analysis to take into account the load bursts. Finally, a general approach is presented that, starting from the concept of system capacity, can help system managers to opportunely set the data center parameters under different working conditions.

A Stochastic Model to Investigate Data CenterPerformance and QoS in Iaas CloudComputing Systems

Cloud data center management is a key problem due to the numerous and heterogeneous strategies that can be applied, ranging from the VM placement to the federation with other clouds. Performance evaluation of Cloud Computing infrastructures is required to predict and quantify the cost-benefit of a strategy portfolio and the corresponding Quality of Service (QoS) experienced by users. Such analyses are not feasible by simulation or on-thefield experimentation, due to the great number of parameters that have to be investigated. In this paper, we present an analytical model, based on Stochastic Reward Nets (SRNs), that is both scalable to model systems composed of thousands of resources and flexible to represent different policies and cloud-specific strategies. Several performance metrics are defined and evaluated to analyze the behavior of a Cloud data center: utilization, availability, waiting time, and responsiveness. A resiliency analysis is also provided to take into account load bursts. Finally, a general approach is presented that, starting from the concept of system capacity, can help system managers to opportunely set the data center parameters under different working conditions.

A Stochastic Model to Investigate Data Center Performance and QoS in IaaS Cloud Computing Systems

Cloud data center management is a key problem due to the numerous and heterogeneous strategies that can be applied, ranging from the VM placement to the federation with other clouds. Performance evaluation of cloud computing infrastructures is required to predict and quantify the cost-benefit of a strategy portfolio and the corresponding quality of service (QoS) experienced by users. Such analyses are not feasible by simulation or on-the-field experimentation, due to the great number of parameters that have to be investigated. In this paper, we present an analytical model, based on stochastic reward nets (SRNs), that is both scalable to model systems composed of thousands of resources and flexible to represent different policies and cloud-specific strategies. Several performance metrics are defined and evaluated to analyze the behavior of a cloud data center: utilization, availability, waiting time, and responsiveness. A resiliency analysis is also provided to take into account load bursts. Finally, a general approach is presented that, starting from the concept of system capacity, can help system managers to opportunely set the data center parameters under different working conditions.

Temperature Monitoring and CFD Analysis of Data Centre

This paper presents a computational fluid dynamics (CFD) analysis of airflow, temperature and pressure distribution of the data centre located at CQUniversity, Rockhampton Campus, Australia. The data centre has been modelled and analyzed using CFD code ‘Fluent’ to study the effectiveness of thermal cooling within the data centre. CoolSim- software was used for designing a graphical user interface (GUI) that allows data centre components to be positioned, sized and characterized in a plan view. The CFD model based on thermal mass and energy balance principles was used to simulate the data centre for predicting the energy consumption, pressure and temperature in a data centre. The effect of various data center parameters on the temperature distribution and the flow field was investigated. The parametric and optimization techniques were used to determine the best possible layout for cooling strategies. The simulation results predict the high temperature zone within the computer rack in the data centre, and provide a detailed 3D analysis of the movement of cold air through the data centre. The results also provide the performance analysis of computer room air conditionings (CRACs) with detailed rack-by-rack inlet and exit temperatures and 3D thermal mapping of the data centre and racks highlighting trouble areas. The model developed is capable of evaluating the airflow rates and thermal loads for optimizing and designing a new or existing data centre.