Disk scheduling for mixed-media workloads in a multimedia server
Peter Triantafillou1998, Proceedings of the sixth ACM international conference on Multimedia - MULTIMEDIA '98
Sign up for access to the world's latest research
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Abstract
AI
AI
Multimedia storage servers experience mixed workloads encompassing both continuous data requests (like audio and video) and discrete data requests (such as text or images). Traditional disk scheduling algorithms primarily focus on continuous requests, often neglecting discrete requests, which can result in performance inconsistencies. This paper introduces novel disk scheduling algorithms designed to address these mixed workloads, aiming to optimize performance for both request types. A simulation-based testbed is utilized to evaluate these algorithms' effectiveness, illustrating their superior performance characteristics compared to existing approaches.
Related papers
Scalable multimedia disk scheduling
Proceedings. 20th International Conference on Data Engineering, 2004
A new multimedia disk scheduling algorithm, termed Cascaded-SFC, is presented. The Cascaded-SFC multimedia disk scheduler is applicable in environments where multimedia data requests arrive with different quality of service (QoS) requirements such as real-time deadline and user priority. Previous work on disk scheduling has focused on optimizing the seek times and/or meeting the real-time deadlines. The Cascaded-SFC disk scheduler provides a unified framework for multimedia disk scheduling that scales with the number of scheduling parameters. The general idea is based on modeling the multimedia disk requests as points in multiple multi-dimensional sub-spaces, where each of the dimensions represents one of the parameters (e.g., one dimension represents the request deadline, another represents the disk cylinder number, and a third dimension represents the priority of the request, etc.). Each multi-dimensional sub-space represents a subset of the QoS parameters that share some common scheduling characteristics. Then the multimedia disk scheduling problem reduces to the problem of finding a linear order to traverse the multi-dimensional points in each sub-space. Multiple space-filling curves are selected to fit the scheduling needs of the QoS parameters in each sub-space. The orders in each sub-space are integrated in a cascaded way to provide a total order for the whole space. Comprehensive experiments demonstrate the efficiency and scalability of the Cascaded-SFC disk scheduling algorithm over other disk schedulers.
A continuum of disk scheduling algorithms
ACM Transactions on Computer Systems, 1987
A continuum of disk scheduling algorithms, V(R), having endpoints V(0) = SSTF and V(1) = SCAN, is defined. V(R) maintains a current SCAN direction (in or out) and services next the request with the smallest effective distance. The effective distance of a request that lies in the current direction is its physical distance (in cylinders) from the read/write head. The effective distance of a request in the opposite direction is its physical distance plus R X (total number of cylinders on the disk). By use of simulation methods, it is shown that this definitional continuum also provides a continuum in performance, both with respect to the mean and with respect to the standard deviation of request waiting time. For objective functions that are linear combinations of the two measures, pw + kuw, intermediate points of the continuum are seen to provide performance uniformly superior to both SSTF and SCAN. A method of implementing V(R) and the results of its experimental use in a real system are presented.
Major Half Served First (MHSF) Disk Scheduling Algorithm
I/O performance has been improved by proper scheduling of disk accesses since the time movable head disk came into existence. Disk scheduling is the process of carefully examining the pending requests to determine the most efficient way to service the pending requests. Scheduling algorithms generally concentrate on reducing seek times for a set of requests, because seek times tend to be an order of magnitude greater than latency times. Some important scheduling algorithms are First-Come-First-Served (FCFS), Shortest Seek Time First (SSTF), SCAN, Circular Scan (C-SCAN) and LOOK. This paper proposes a new disk scheduling algorithm called Major Half Served First (MHSF). Simulation results show that using MHSF the service is fast and seek time has been reduced drastically.
A Novel Deadline Driven Disk Scheduling Algorithm for Multi-Priority Multimedia Objects
International Conference on Data Engineering, 2000
In this paper we introduce a new deadline driven diskscheduling algorithm designed for multimedia servers.The proposed algorithm supports real time requestswith multiple priorities, e.g., those for di#erent objectclasses in digital library applications. The proposedalgorithm enhances utilization of disk bandwidthby #a# maintaining one queue for all requests, and #b#optimizing the seek time. Prior schemes, collectivelytermed #Multi-queue schemes", maintain a separatequeue for
Clustered scheduling algorithms for mixed-media disk workloads in a multimedia server
2003
Divisible load scenarios occur in modern media server applications since most multimedia applications typically require access to continuous and discrete data. A high performance Continuous Media (CM) server greatly depends on the ability of its disk IO subsystem to serve both types of workloads efficiently. Disk scheduling algorithms for mixed media workloads, although they play a central role in this task, have been overlooked by related research efforts. These algorithms must satisfy several stringent performance goals, such as achieving low response time and ensuring fairness, for the discrete-data workload, while at the same time guaranteeing the uninterrupted delivery of continuous data, for the continuous-data workload. The focus of this paper is on disk scheduling algorithms for mixed media workloads in a multimedia information server. We propose novel algorithms, present a taxonomy of relevant algorithms, and study their performance through experimentation. Our results show that our algorithms offer drastic improvements in discrete request average response times, are fair, serve continuous requests without interruptions, and that the disk technology trends are such that the expected performance benefits can be even greater in the future.
Disk scheduling for variable-rate data streams
Lecture Notes in Computer Science, 1997
We describe three disk scheduling algorithms that can be used in a multimedia server for sustaining a number of heterogeneous variable-rate data streams. A data stream is supported by repeatedly fetching a block of data from the storage device and storing it in a corresponding buffer. For each of the disk scheduling algorithms we give necessary and sufficient conditions for avoiding under-and overflow of the buffers. In addition, the algorithms are compared with respect to buffer requirements as well as average response times.
Efficient guaranteed disk request scheduling with fahrrad
Proceedings of the 3rd ACM SIGOPS/EuroSys European Conference on Computer Systems 2008 - Eurosys '08, 2008
Guaranteed I/O performance is needed for a variety of applications ranging from real-time data collection to desktop multimedia to large-scale scientific simulations. Reservations on throughput, the standard measure of disk performance, fail to effectively manage disk performance due to the orders of magnitude difference between best-, average-, and worst-case response times, allowing reservation of less than 0.01% of the achievable bandwidth. We show that by reserving disk resources in terms of utilization it is possible to create a disk scheduler that supports reservation of nearly 100% of the disk resources, provides arbitrarily hard or soft guarantees depending upon application needs, and yields efficiency as good or better than best-effort disk schedulers tuned for performance. We present the architecture of our scheduler, prove the correctness of its algorithms, and provide results demonstrating its effectiveness.
A New Heuristic Disk Scheduling Algorithm
International Journal of Scientific & Technology Research, 2013
Since the invention of the movable head disk, people have improved I/O performance by intelligent scheduling of disk accesses. Processor speed and memory capacity are increasing several times faster than disk speed. This disparity suggests that disk I/O performance w ill become an important bottleneck .Methods are needed for using disks more efficiently. Past analysis of disk scheduling algorithms has largely been experimental and little attempt has been made to develop algorithms w ith provable performance guarantees. Disk performance management is an increasingly important aspect of operating system research and development. In this paper a new disk scheduling algorithm has been proposed to reduce the number of movement of head. It is observed that in existing scheduling algorithms the number of head movement is high. But we proposed a new real- time disk scheduling algorithm that reduces the head movement therefore it maximizes throughput for modern storage devices.
An improved parallel disk scheduling algorithm
Proceedings. Fifth International Conference on High Performance Computing (Cat. No. 98EX238)
We address the problems of prefetching and I/O scheduling for read-once reference strings in a parallel I/O system. Read-once reference strings, in which each block is accessed exactly once, arise naturally in applications like databases and video retrieval. Using the standard parallel disk model with £ disks and a shared I/O buffer of size ¤ , we present a novel algorithm, Red-Black Prefetching (RBP), for parallel I/O scheduling. The number of parallel I/Os performed by RBP is within O(£ ¦ ¥ § ©) of the minimum possible. Algorithm RBP is easy to implement and requires computation time linear in the length of the reference string. Through simulation experiments we validated the benefits of RBP over simple greedy prefetching.
Scalable QoS-aware disk-scheduling
Proceedings International Database Engineering and Applications Symposium, 2002
A new quality of service (QoS) aware disk scheduling algorithm is presented. It is applicable in environments where data requests arrive with different QoS requirements such as real-time deadline, and user priority. Previous work on disk scheduling has focused on optimizing the seek times and/or meeting the real-time deadlines. A unified framework for QoS disk scheduling is presented that scales with the number of scheduling parameters. The general idea is based on modeling the disk scheduler requests as points in the multi-dimensional space, where each of the dimensions represents one of the parameters (e.g., one dimension represents the request deadline, another represents the disk cylinder number, and a third dimension represents the priority of the request, etc.). Then the disk scheduling problem reduces to the problem of finding a linear order to traverse these multi-dimensional points. Space-filling curves are adopted to define a linear order for sorting and scheduling objects that lie in the multi-dimensional space. This generalizes the one-dimensional disk scheduling algorithms (e.g., EDF, SATF, FIFO). Several techniques are presented to show how a QoS-aware disk scheduler deals with the progressive arrival of requests over time. Simulation experiments are presented to show a comparison of the alternative techniques and to demonstrate the scalability of the proposed QoSaware disk scheduling algorithm over other traditional approaches.
Related papers
A Comparative Analysis of Disk Scheduling Algorithms
International Journal for Research in Applied Sciences and Biotechnology, 2021
In an operating system, disk scheduling is the process of managing the I/O request to the secondary storage devices such as hard disk. The speed of the processor and primary memory has increased in a rapid way than the secondary storage. Seek time is the important factor in an operating system to get the best access time. For the better performance, speedy servicing of I/O request for secondary memory is very important. The goal of the disk-scheduling algorithm is to minimize the response time and maximize throughput of the system. This work analyzed and compared various basic disk scheduling techniques like First Come First Serve (FCFS), Shortest Seek Time First (SSTF), SCAN, LOOK, Circular SCAN (C-SCAN) and Circular LOOK (C-LOOK) along with the corresponding seek time. From the comparative analysis, the result show that C-LOOK algorithm give the least head movement and seek time in different cases as compared to other algorithm. Therefore, it maximizes the throughput for the storage devices.
A hierarchical disk scheduler for multimedia systems
Future Generation Computer Systems, 2003
An integrated storage platform for open systems should be able to meet the requirements of deterministic applications, multimedia systems, and traditional best-effort applications. It should also provide a scheduling mechanism fitting all those types of applications. In this paper, we propose a two-level hierarchical disk scheduling scheme, named 2-Q, which can guarantee deterministic deadlines, maximize the number of statistic real-time streams processed by the disk system, and minimize the average latency for best-effort requests. The upper level of the scheduling architecture, server level, is divided into three queues: deterministic, statistic, and best-effort requests. Each server may have its own scheduling algorithm. The lower level, disk driver, chooses the ready streams using its own scheduling criteria. We also propose an adaptive admission control algorithm relying on worst and average values of disk server utilization. Only streams satisfying the admission algorithm criteria are accepted for further processing by the disk server. The solution is extended to a parallel disk system by using a third hierarchical level, named meta-scheduler, briefly described in the paper. The performance evaluations demonstrate that our scheduling architecture is adequated for handling stream sets with different deterministic, statistic, or best-effort requirements.
Disk Scheduling
IJARCSMS
To service a request, a disk system requires that the head be moved to the desired track, then a wait for latency and finally the transfer of data.
CSched : Real-time disk scheduling with concurrent I/O request
2011
We present a new real-time disk scheduling algorithm, Concurrent Scheduler or CSched, which maximizes throughput for modern storage devices while providing real-time access guarantees, with computational costs of O(log n). To maximize performance it ensures request concurrency at the device and maximizes the depth of a new Limited Cyclical SCAN (L-CSCAN) queue that optimizes the request sequence sent to the device. For realtime requests there is an additional SCAN-EDF queue in front of the L-CSCAN queue to absorb bursts of realtime requests until they can be drained to the L-CSCAN queue. The real-time guarantees are provided by managing the worst-case latency at each stage of the pipeline: SCAN-EDF, L-CSCAN, and device. CSched is configured by the tuple {λ, σ, δ, τ(r), N}, where λ and σ are the minimal initial slack time and workload burstiness and are properties of the workload, and where δ, τ(r), and N are the deviceworst-case latency, worst-case throughput rate time for a request, and maximal number of concurrent requests, and are experimentally determined properties of the storage device. An experimental evaluation of CSched shows that given sufficient initial slack time, the system throughput performance costs of providing real-time guarantees are negligible.
S-LOOK: A Preemptive Disk Scheduling Algorithm for Offline and Online Environments
Кеу wordsscheduling algorithms, scheduling simulator.
Real-Time Disk Scheduling for Multimedia Applications with Deadline-Modification-Scan Scheme
Real-time Systems - RTS, 2000
Real-timedisk scheduling (RTDS) is important for time-critical multimediaapplications. In conventional approaches of real-time disk scheduling,such as SCAN-EDF, the seek-optimizing SCAN scheme is appliedto reduce the disk service time of the real-time EDF schedule.Since only tasks with the same deadline are seek-optimized, theobtained improvement of SCAN-EDF is limited. In this paper, basedon the maximum-scannable-group (MSG) concept, a deadline-modification-scan (DMS or DM-SCAN) algorithm is proposed.Our algorithm uses MSG to automatically decide the suitable taskgroups for seek-optimizing. Based on the MSG concept, we repeatedlyapply DMS to further increase disk throughput to support moretasks. We have implemented the proposed algorithm on UnixWare2.01. The appropriate problem behaviors and parameter valuesto yield good solutions are investigated. Experiments show thatDMS is better than conventional approaches in both the obtaineddisk throughput and the supported tasks. Moreover, ...
Simulation and Performance Comparison of Four Disk Schedulingalgorithms
TENCON 2000. Proceedings, 2000
Hard disks are being used to store huge informatioddata in all modem computers. Disk drives must provide faster access time in order to optimize speed of I/O operations. In multitasking system with many processes, disk performance can be improved by incorporating a scheduling algorithm for maintaining several pending requests in the disk queue. This paper describes development of a simulator which uses four disk scheduling algorithms (FCFS, SSTF, LOOK for both upward and downward direction, and C-LOOK) to measure their performance in terms of total head movement. Five different types of test samples, containing request tracks from 8 to 50, have been used to obtain simulation results. Developed simulator runs successfully
Disk scheduling with shortest cumulative access time first algorithms
Turkish J. Electr. Eng. Comput. Sci., 2017
A new class of scheduling algorithms is proposed for disk drive scheduling. As opposed to choosing the request with the shortest access time in conventional shortest access time first (SATF) algorithms, we choose an ordered sequence of pending I/O requests at the scheduling instant with the shortest cumulative access time. Additionally, we introduce flexibility for forthcoming requests to alter the chosen sequence. Simulation results are provided to validate the effectiveness of the proposed disk scheduler. Throughput gains of 3% and above are shown to be attainable, although this occurs at the expense of increased computational complexity.
Design and Performance Evaluation of an Optimized Disk Scheduling Algorithm (ODSA)
International Journal of …, 2012
Management of disk scheduling is a very important aspect of operating system. Performance of the disk scheduling completely depends on how efficient is the scheduling algorithm to allocate services to the request in a better manner. Many algorithms (FIFO, SSTF, SCAN, C-SCAN, LOOK, etc.) are developed in the recent years in order to optimize the system disk I/O performance. By reducing the average seek time and transfer time, we can improve the performance of disk I/O operation. In our proposed algorithm, Optimize Disk Scheduling Algorithm (ODSA) is taking less average seek time and transfer time as compare to other disk scheduling algorithms (FIFO, SSTF, SCAN, C-SCAN, LOOK, etc.), which enhances the efficiency of the disk performance in a better manner.
Real-time disk scheduling algorithm allowing concurrent I/O requests
2009
real-time disk scheduling, storage systems, operating systems, RAID We present a real-time disk scheduling algorithm, Concurrent DS-SCAN (CDS-SCAN), which maximizes throughput for modern storage devices by allowing concurrent I/O requests at the device whenever possible. Past real-time disk scheduling algorithms allowed a single request at a time to go to the storage device, which dramatically reduces the utilization and throughput for modern storage devices, such as RAID arrays and disks with efficient positional-aware scheduling algorithms. We extended the DS-SCAN algorithm so that it can properly account for multiple outstanding I/O requests and guarantee real-time constraints for both outstanding and pending real-time requests. We demonstrate CDS-SCAN's performance on a storage array.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.