Data centers are often under-utilized due to over-provisioning as well time-varying resource demands of typical enterprise applications. One approach increase utilization is consolidate applications in a shared infrastructure using virtualization. Meeting application-level quality service (QoS) goals becomes challenge consolidated environment application needs differ. Furthermore, for multi-tier applications, the amount resources needed achieve their QoS might be different at each tier and...

Virtualized data centers enable sharing of resources among hosted applications. However, it is difficult to satisfy service-level objectives(SLOs) applications on shared infrastructure, as application workloads and resource consumption patterns change over time. In this paper, we present AutoControl, a control system that automatically adapts dynamic workload changes achieve SLOs. AutoControl combination an online model estimator novel multi-input, multi-output (MIMO) controller. The...

Recent advances in hardware and software virtualization offer unprecedented management capabilities for the mapping of virtual resources to physical resources. It is highly desirable further create a "service hosting abstraction" that allows application owners focus on service level objectives (SLOs) their applications. This calls resource solution achieves SLOs many applications response changing data center conditions hides complexity from both operators. In this paper, we describe an...

Data centers are often under-utilized due to over-provisioning as well time-varying resource demands of typical enterprise applications. One approach increase utilization is consolidate applications in a shared infrastructure using virtualization. Meeting application-level quality service (QoS) goals becomes challenge consolidated environment application needs differ. Furthermore, for multi-tier applications, the amount resources needed achieve their QoS might be different at each tier and...

The multi-pulse excitation model provides a method for producing natural-sounding speech at medium to low bit rates. Multi-pulse analysis obtains the all-pole filter by minimizing spectrally-weighted mean-squared error between original and synthetic signals. Although high quality around 10 kbits/sec, suffers if rate is lowered. In this paper, we focus on problems encountered in attempting maintain while synthesizing using lower

Data centers contain IT, power and cooling infrastructures, each of which is typically managed independently. In this paper, we propose a holistic approach that couples the management infrastructures to improve efficiency data center operations. Our considers application performance management, dynamic workload migration/consolidation, control "right-provision" computing, resources for given workload. We have implemented prototype virtualized environments conducted experiments in production...

Feedback mechanisms can help today's increasingly complex computer systems adapt to changes in workloads or operating conditions. Control theory offers a principled way for designing feedback loops deal with unpredictable changes, uncertainties, and disturbances systems. We provide an overview of the joint research at HP Labs University Michigan past few years, where control was applied automated resource service level management data centers. highlight key benefits control-theoretic...

In this paper, we describe the design of online feedback control algorithms to dynamically adjust entitlement values for a resource container on server shared by multiple applications. The goal is determine minimum level such that its hosted application achieves desired performance levels. Classic theory used both model identification and controller design. Specific implementation issues affect closed-loop system are discussed. A self-tuning adaptive also presented handle limited variations...

A new motion compensation technique using a window which satisfies the perfect reconstruction condition is proposed. The conventional rectangular blocks often gives discontinuities between neighboring in predicted image. proposed method based on operation to data overlaps an area of block. Computer simulation carried out MPEG video coding algorithm evaluate method. performance better than terms mean square error, and large improvement can be obtained at block boundaries. This smooth image...

Although the multipulse model is conceptually simple, problem of locating pulses computationally complex. The authors discuss basic and describe a procedure to compute excitation with optimally adjusted amplitudes. algorithm provides framework for computing varying degrees optimization computational complexity. find that speech quality depends on pulse rate. They also same quality, female requires higher rate than male speech. pitch dependence can be reduced improved high-pitched speakers by...

Technologies like RDMA and Gen-Z, which give access to memory outside the box, are gaining in popularity. These technologies provide abstraction of far memory, where is attached network can be accessed by remote processors without mediation a local processor. Unfortunately, hard use because existing data structures mismatched it. We argue that we need new for borrowing techniques from concurrent distributed systems. examine requirements these show how realize them using simple hardware extensions.

It is challenging to reduce resource over-provisioning for enterprise applications while maintaining service level objectives (SLOs) due their time-varying and stochastic workloads. In this paper, we study the effect of prediction on dynamic allocation virtualized servers running applications. We present predictive controllers using three different algorithms based a standard auto-regressive (AR) model, combined ANOVA-AR as well multi-pulse (MP) model. compare properties with an adaptive...

Today's shared hosting platforms often employ virtualization to allow multiple enterprise applications with time-varying resource demands share a common infrastructure in order improve utilization. Meeting application-level quality of service (QoS) goals becomes challenge such an environment as have multi-tier architecture and complex interactions dependencies among individual tiers. In addition, when the overloaded, appropriate control needs be performed at these tiers coordinated fashion...

Virtualization and consolidation of IT resources have created a need for more effective workload management tools, one that dynamically controls resource allocation to hosted application achieve quality service (QoS) goals. These goals can in turn be driven by the utility service, typically based on application's level agreement (SLA) as well cost allocated. In this paper, we build our earlier work dynamic CPU applications shared servers, present feedback control system consisting two nested...

Managing application-level performance for multitier applications in virtualized server environments is challenging because the are distributed across multiple virtual machines, and workloads dynamic their intensity transaction mix resulting time-varying resource demands. In this paper, we present AppRAISE, a system that manages of multi-tier by dynamically resizing machines hosting applications. We extend traditional queuing model to represent application environments, where machine...

An experimental open, cloud-based platform for large-scale, low-cost delivery of healthcare applications enables broader use patient-centric management electronic health records and facilitates the secure seamless sharing EHRs among stakeholders within a system.

We present Dinomo, a novel key-value store for disaggregated persistent memory (DPM). Dinomo is the first DPM that simultaneously achieves high common-case performance, scalability, and lightweight online reconfiguration. observe previously proposed stores had architectural limitations prevent them from achieving all three goals simultaneously. uses combination of techniques such as ownership partitioning, adaptive caching, selective replication, lock-free log-free indexing to achieve these...

The fast growth of cloud service offerings has attracted more enterprises to migrate their IT applications into cloud. Nonetheless, complex enterprise user requirements, especially interdependent relations across them, raise new challenges selection. In addition, a major concern for these is ensuring compliance with policies on the use services. this paper, we present systematic framework, based formal verification and constraint solving techniques, help tackle problems when adopting...

Provisioning storage in disk arrays is a difficult problem because many applications with different workload characteristics and priorities share resources provided by the array. Currently, statically partitioned, leading to choices between over-provisioning meet peak demands resource sharing efficiency targets. In this paper, we present Maestro, feedback controller that can manage on large provide performance differentiation among multiple applications. Maestro monitors of each application...