Richard P. King

Web services on demand: WSLA-driven automated management

In this paper we describe a framework for providing customers of Web services differentiated levels of service through the use of automated management and service level agreements (SLAs). The framework comprises the Web Service Level Agreement (WSLA) language, designed to specify SLAs in a flexible and individualized way, a system to provision resources based on service level objectives, a workload management system that prioritizes requests according to the associated SLAs, and a system to monitor compliance with the SLA. This framework was implemented as the utility computing services part of the IBM Emerging Technologies Tool Kit, which is publicly available on the IBM alphaWorksTM Web site.

Network dispatcher: a connection router for scalable Internet services

Abstract Network Dispatcher (ND) is a TCP connection router that supports load sharing across several TCP servers. Prototypes of Network Dispatcher were used to support several large scale high-load Web sites. Network Dispatcher provides a fast IP packet-forwarding kernel-extension to the TCP IP stack. Load sharing is supported by a user-level manager process that monitors the load on the servers and controls the connection allocation algorithm in the kernel extension. This paper describes the design of Network Dispatcher, outlines Network Dispatchers performance in the context of http traffic, and presents several of its features including high-availability, support for WANs, and client affinity.

Design, implementation, and evaluation of the linear road bnchmark on the stream processing core

Stream processing applications have recently gained significant attention in the networking and database community. At the core of these applications is a stream processing engine that performs resource allocation and management to support continuous tracking of queries over collections of physically-distributed and rapidly-updating data streams. While numerous stream processing systems exist, there has been little work on understanding the performance characteristics of these applications in a distributed setup. In this paper, we examine the performance bottlenecks of streaming data applications, in particular the Linear Road stream data management benchmark, in achieving good performance in large-scale distributed environments, using the Stream Processing Core (SPC), a stream processing middleware we have developed. First, we present the design and implementation of the Linear Road benchmark on the SPC middleware. SPC has been designed to scale to tens of thousands of processing nodes, while supporting concurrent applications and multiple simultaneous queries. Second, we identify the main performance bottlenecks in the Linear Road application in achieving scalability and low query response latency. Our results show that data locality, buffer capacity, physical allocation of processing elements to infrastructure nodes, and packaging for transporting streamed data are important factors in achieving good application performance. Though we evaluate our system primarily for the Linear Road application, we believe it also provides useful insights into the overall system behavior for supporting other distributed and large-scale continuous streaming data applications. Finally, we examine how SPC can be used and tuned to enable a very efficient implementation of the Linear Road application in a distributed environment.

A service level agreement language for dynamic electronic services

This paper proposes a novel language for Service Level Agreements (SLAs) for dynamic and spontaneous electronic services. In a cross-organizational setting, it is important for customers of a service to obtain, monitor and enforce quality of service (QoS) guarantees by service providers, usually expressed in the form of SLAs. Since the supervision and management of SLAs and the provisioning of corresponding systems should be automated for economic reasons, we need a formal language to define an SLA. If moreover, providers and customers want to sign custom-made SLAs, the SLA language, correspondingly, must provide a large degree of flexibility. The SLA language described in this paper aims at providing the needed flexibility by means of an XML-based representation and a runtime system for SLAs. Using this language, parties to a SLA can describe how parameters are measured and computed from raw metrics, the guarantees they want with respect to those parameters and the involvement of third parties to, e.g., verify independently SLA compliance.

SPC: a distributed, scalable platform for data mining

The Stream Processing Core (SPC) is distributed stream processing middleware designed to support applications that extract information from a large number of digital data streams. In this paper, we describe the SPC programming model which, to the best of our knowledge, is the first to support stream-mining applications using a subscription-like model for specifying stream connections as well as to provide support for non-relational operators. This enables stream-mining applications to tap into, analyze and track an ever-changing array of data streams which may contain information relevant to the streaming-queries placed on it. We describe the design, implementation, and experimental evaluation of the SPC distributed middleware, which deploys applications on to the running system in an incremental fashion, making stream connections as required. Using micro-benchmarks and a representative large-scale synthetic stream-mining application, we evaluate the performance of the control and data paths of the SPC middleware.

Management of a remote backup copy for disaster recovery

A remote backup database system tracks the state of a primary system, taking over transaction processing when disaster hits the primary site. The primary and backup sites are physically isolated so that failures at one site are unlikely to propogate to the other. For correctness, the execution schedule at the backup must be equivalent to that at the primary. When the primary and backup sites contain a single processor, it is easy to achieve this property. However, this is harder to do when each site contains multiple processors and sites are connected via multiple communication lines. We present an efficient transaction processing mechanism for multiprocessor systems that guarantees this and other important properties. We also present a database initialization algorithm that copies the database to a backup site while transactions are being processed.

Disk arm movement in anticipation of future requests

When a disk drives access arm is idle, it may not be at the ideal location. In anticipation of future requests, movement to some other location may be advantageous. The effectiveness of anticipatory disk arm movement is explored. Various operating conditions are considered, and the reduction in seek distances and request response times is determined for them. Suppose that successive requests are independent and uniformly distributed. By bringing the arm to the middle of its range of motion when it is idle, the expected seek distance can be reduced by 25 percent. Nonlinearity in time versus distance can whittle that 25 percent reduction down to a 13 percent reduction in seek time. Nonuniformity in request location, nonPoisson arrival processes, and high arrival rates can whittle the reduction down to nothing. However, techniques are discussed that maximize those savings that are still possible under those circumstances. Various systems with multiple arms are analyzed. Usually, it is best to spread out the arms over the disk area. The both arms should be brought to the middle.

A Service Level Agreement Language for Dynamic Electronic Services

This paper proposes a novel language for Service Level Agreements (SLAs) for dynamic and spontaneous electronic services. In a cross-organizational setting, it is important for customers of a service to obtain, monitor and enforce quality of service (QoS) guarantees by service providers, usually expressed in the form of SLAs. Since the supervision and management of SLAs and the provisioning of corresponding systems should be automated for economic reasons, we need a formal language to define an SLA. If, moreover, providers and customers want to sign custom-made SLAs, the SLA language, correspondingly, must provide a large degree of flexibility.The SLA language described in this paper aims at providing the needed flexibility by means of an XML-based representation and a runtime system for SLAs. Using this language, parties to an SLA can describe how parameters are measured and computed from raw metrics, the guarantees they want with respect to those parameters and the involvement of third parties to, e.g., independently verify SLA compliance.

The Coyote Project: Framework for Multi-party E-Commerce

The Internet provides the opportunity for quickly setting up deals between businesses for promoting each others products, and to jointly offer new services. Specification and enforcement of such deals stretch traditional transaction processing concepts in several directions since they involve independent businesses with their own internal processes. First, the greater variability in response time in business to business interaction creates a need for asynchronous and event-driven processing, in which correct handling of reissued and cancelled requests is critical. Second, a new transaction processing paradigm is required that supports different views of a unit of business for all participants, i.e., service providers as well as end consumers. Between any two interacting parties, there may be several related interactions dispersed in time, creating a long running conversation. This paper describes our approach (Coyote) to solving these problems including use of a service contract for specifying the rules of interaction across businesses, and directly generating code for enforcement of the contract. We finally describe the architecture and a prototype of a system which implements the Coyote concepts.

Placement of multimedia blocks on zoned disks

We examine the impact on optimal data placement schemes for random retrieval due to a new trend in disk technology, namely zoned magnetic disks, which record data with a constant linear density and have a constant angular velocity. This modifies the optimal data placement that minimizes response time, since now the data transfer time varies with block placement whereas the transfer time and cylinder capacity was constant for traditional magnetic disks. We focus on the optimal placement of video objects stored in multimedia servers that consist of fixed-size blocks of constant-bit-rate encoded data or varying-size blocks of variable-bit-rate encoded data. The differences in popularity and access rates among these objects allow minimization of seek times by varying the data placement. We describe an optimal placement of fixed-size blocks on zoned disks that minimizes response time. Video objects are accessed by streams that playout the blocks at a given rate, requiring disk access by a deadline in order to avoid glitches in playout. For the placement of varying-size blocks on zoned disks, we study a scheme that minimizes the probability that a block access misses its deadline. This is done by adjusting the value representing the popularity of blocks to take into account both block size and the relative importance of seek time and transfer time using a weighting factor. We quantify the improvements in disk throughput for a given block-deadline miss probability.