Banu Özden

Cyclic association rules

We study the problem of discovering association rules that display regular cyclic variation over time. For example, if we compute association rules over monthly sales data, we may observe seasonal variation where certain rules are true at approximately the same month each year. Similarly, association rules can also display regular hourly, daily, weekly, etc., variation that is cyclical in nature. We demonstrate that existing methods cannot be naively extended to solve this problem of cyclic association rules. We then present two new algorithms for discovering such rules. The first one, which we call the sequential algorithm, treats association rules and cycles more or less independently. By studying the interaction between association rules and time, we devise a new technique called cycle pruning, which reduces the amount of time needed to find cyclic association rules. The second algorithm, which we call the interleaved algorithm, uses cycle pruning and other optimization techniques for discovering cyclic association rules. We demonstrate the effectiveness of the interleaved algorithm through a series of experiments. These experiments show that the interleaved algorithm can yield significant performance benefits when compared to the sequential algorithm. Performance improvements range from 5% to several hundred percent.

Disk scheduling with quality of service guarantees

The paper introduces YFQ, a new disk scheduling algorithm that allows applications to set aside for exclusive use portions of the disk bandwidth. We implemented YFQ as part of the Eclipse/BSD operating system, which is derived from FreeBSD, a version of 4.4 BSD Unix. YFQs disk bandwidth reservations can guarantee file accesses with high throughput, low delay, and good fairness. Such quality of service (QoS) guarantees to individual applications unfortunately can also hinder global disk scheduling optimizations. We propose and evaluate several disk scheduling enhancements that promote global optimizations and give to YFQ aggregate disk throughput approaching that of FreeBSDs conventional disk scheduler, which does not provide QoS guarantees. We believe that our enhancements may be helpful also in other disk scheduling algorithms.

Buffer replacement algorithms for multimedia storage systems

In a disk-based storage system, a buffer cache is used to reduce the number of disk I/Os. The buffer manager is responsible for buffer replacement to free memory in order to accommodate new data blocks from the disk. An optimal buffer replacement algorithm is one which yields the lowest number of cache misses and, as a result, the lowest number of disk I/Os. In general, optimal buffer replacement algorithms require future knowledge, and therefore cannot be realized. Thus, most storage systems use approximation algorithms such as the least-recently-used (LRU) and most-recently-used (MRU) buffer replacement algorithms. In this paper, we show that both of these algorithms yield poor performance when they are used in storage systems that provide support for continuous media data. We present two new buffer replacement algorithms-BASIC and DISTANCE-which reduce the cache misses by up to 30% as compared to LRU and MRU. Furthermore, in the simulation experiments we conducted, our new algorithms resulted in at most a 9.7% increase in cache misses as compared to the optimal algorithm when the videos are sufficiently long (i.e. longer than 30 minutes). Moreover, we show that the DISTANCE scheme incurs an overhead which is comparable to those of the LRU and MRU schemes. Our conclusion is that DISTANCE is a very suitable candidate for a buffer replacement scheme in storage systems that deal with continuous media data.

Disk striping in video server environments

A growing number of applications need access to video data stored in digital form on secondary storage devices (e.g., video-on-demand, multimedia messaging). As a result, video servers that are responsible for the storage and retrieval, at fixed rates, of hundreds of videos from disks are becoming increasingly important. Since video data tends to be voluminous, several disks are usually used in order to store the videos. A challenge is to devise schemes for the storage and retrieval of videos that distribute the workload evenly across disks, reduce the cost of the server and at the same time, provide good response times to client requests for video data. We present schemes that are based on striping videos (fine-grained as well as coarse-grained) across disks in order to effectively utilize disk bandwidth. The proposed schemes are starvation-free and provide good response times to client requests. For the schemes, we show how an optimal-cost server architecture can be determined if data for a certain prespecified number of videos is to be concurrently retrieved. Finally, through extensive simulations, we compare the performance of the various schemes.

Fair queuing for aggregated multiple links

Provisioning of a shared server with guarantees is an important scheduling task that has led to significant work in a number of areas including link scheduling. Fair Queuing algorithms provide a method for proportionally sharing a single server among competing flows, however, they do not address the problem of sharing multiple servers. Multiserver systems arise in a number of applications including link aggregation, multiprocessors and multi-path storage I/O. In this paper we introduce a new service discipline for multi-server systems that provides guarantees for competing flows. We prove that this new service discipline is a close approximation of the idealized Generalized Processor Sharing (GPS) discipline. We calculate its maximum packet delay and service discrepancy with respect to GPS. We also discuss its relevance to several applications, in particular, Ethernet link aggregation.

On the design of a low-cost video-on-demand storage system

Recent advances in storage technology, coupled with the dramatic increase in the bandwidth of networks, now make it possible to provide “video-on-demand” service to viewers. A video-on-demand server is a computer system that stores videos in compressed digital form and provides support for various portions of compressed video data to be accessed and transmitted concurrently. We present a low-cost storage architecture for a video-on-demand server that relies principally on disks. The high bandwidths of disks in conjunction with a clever strategy for striping videos on them enables simultaneous access and transmission of portions of a video, separated by fixed time intervals. We also present schemes for implementing VCR-like functions including fast forward, rewind, and pause, and extend our schemes to the case in which videos have different rate requirements.

Fault-tolerant architectures for continuous media servers

Continuous media servers that provide support for the storage and retrieval of continuous media data (e.g., video, audio) at guaranteed rates are becoming increasingly important. Such servers, typically, rely on several disks to service a large number of clients, and are thus highly susceptible to disk failures. We have developed two fault-tolerant approaches that rely on admission control in order to meet rate guarantees for continuous media requests. The schemes enable data to be retrieved from disks at the required rate even if a certain disk were to fail. For both approaches, we present data placement strategies and admission control algorithms. We also present design techniques for maximizing the number of clients that can be supported by a continuous media server. Finally, through extensive simulations, we demonstrate the effectiveness of our schemes.

A framework for the storage and retrieval of continuous media data

Continuous media applications require a guaranteed transfer rate of data, which conventional storage servers are not designed to provide. The aim of this paper is to provide a general framework for the design of storage servers that deal with both continuous and non-continuous media data. We present several algorithms for the concurrent transfer of continuous media data for multiple requests with different rates. The algorithms provide high throughout by reducing the seek latency time and by eliminating relational latency incurred when accessing data on disks. Each of these algorithms is accompanied by an admission control scheme to restrict the number of concurrent requests being serviced at any given time. We also augment these algorithms to support conventional data accesses without violating the rate guarantees of continuous media data requests. Finally, we extend our algorithms to deal with the newer disks, where transfer rates vary from one track to another. The algorithms presented in this paper are used in Fellini-a storage server for continuous and conventional data being implemented at AT&T Bell Laboratories.

Demand paging for video-on-demand servers

With recent advances in storage and network technology it is now possible to provide video on demand (VOD) service, thereby eliminating the inflexibility inherent in todays broadcast cable systems. A VOD server is a computer system that stores videos in compressed digital form and provides support for the concurrent transmission of different portions of the compressed video data to the various viewers. We present novel demand paging algorithms that provide rate guarantees while utilizing the limited buffer space effectively and eliminating the disk bandwidth limitation. Our schemes, therefore, increase the number of clients that can be serviced concurrently. A VOD server, which is based on our schemes, is currently being implemented at AT&T.

Move-to-rear list scheduling: a new scheduling algorithm for providing QoS guarantees

In order to support multiple real time applications on a sin gle platform the operating system must provide Quality of Service QoS guarantees so that the system resources can be provisioned among applications to achieve desired levels of predictable performance The traditional QoS pa rameters include fairness delay and throughput In this paper we introduce a new QoS criterion called cumulative service The cumulative service criterion relates the total service obtained by a process under a scheduling policy to the ideal service that the process would have accumulated by executing on each resource at a reserved rate We say that a scheuling policy provides a cumulative service guar antee if the performance of the real system di ers from the ideal system by at most a constant amount A cumulative service guarantee is vital for applications e g a continous media le service that require multiple resources and de mand predictable aggregated throughput over all these re sources Existing scheduling algorithms that guarantee tra ditional QoS paramaters do not provide cumulative service guarantees We present a new scheduling algorithm called Move To Rear List Scheduling which provides a cumulative service guarantee as well as the traditional guarantees such as fairness proportional sharing and bounded delay The complexity of MTR LS is O ln n where n is the number