• Journal of the Korea Society of Computer and Information
• /
• v.20 no.10
• /
• pp.91-97
• /
• 2015

In this paper, we propose a MCFQ I/O scheduler that is implemented by modifying the existing Linux CFQ I/O scheduler. MCFQ observes whether the user requested I/O bandwidth weight is well distributed. Based on the I/O bandwidth observation, we improved I/O performance of the existing bandwidth distribution ability by dynamically controlling the I/O time-slice of the virtual machine. The use of SSDs as storage has been increasing dramatically in recent computer systems due to their fast performance and low power usage. As the usage of SSD increases and prices fall, virtualized system administrators can take advantage of SSDs. However, studies on guaranteeing SLA(Service Level Agreement) services when multiple virtual machines share the SSD is still incomplete. In this paper was conducted to improve performance of the bandwidth distribution when multiple virtual machine are sharing a single SSD storage in a virtualized environment. In particular, it was observed that the performance of the bandwidth distribution varied widely when garbage collection occurs in the SSD. In order to reduce performance variance, we add a MoTS(Manager of Time Slice) on existing CFQ I/O scheduler.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.237-245
• /
• 2023

In the cloud service, system resource such as CPU, memory, I/O bandwidth are shared among multiple users. Particularly, in Linux containers environment, I/O bandwidth is distributed in proportion to the weight of each container through the BFQ I/O scheduler. However, since the I/O scheduler can only be applied to conventional block storage devices, it cannot be applied to Zoned Namespace(ZNS) SSD, a new storage interface that has been recently studied. To overcome this limitation, in this paper, we implemented a weighted proportional I/O bandwidth sharing scheme for ZNS SSDs in dm-zoned, which emulates conventional block storage using ZNS SSDs. Each user receives a different amount of budget, which is required to process the user's I/O requests based on the user's weight. If the budget is exhausted I/O requests cannot be processed and requests are queued until the budget replenished. Each budget refill period, the budget is replenished based on the user's weight. In the experiment, as a result, we can confirm that the I/O bandwidth can be distributed on their weight as we expected.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.3
• /
• pp.97-103
• /
• 2020

Linux Cgroups takes a fundamental role for sharing system resources among multiple containers on container-based cloud computing environment. Especially for I/O resource, Linux Cgroups supports a mechanism for sharing I/O bandwidth in proportion to I/O weight. However, the current mechanism of Linux Cgroups using BFQ I/O scheduler seriously degrades the I/O performance with high bandwidth storage device such as NVMe SSDs. In this paper, we proposed a new feedback based I/O bandwidth sharing scheme for Linux Cgroups which allocates I/O credits to containers according to I/O weights and adjusts the amount of credits to performance fluctuation of NVMe SSDs. The proposed scheme is implemented on Linux kernel 5.3 and evaluated. The evaluation results show that it can share the I/O bandwidth among multiple containers proportionally to I/O weights while improving I/O performance more than twice as high as the existing scheme.

• Journal of Korea Multimedia Society
• /
• v.13 no.11
• /
• pp.1594-1609
• /
• 2010

In the case that multiple services are in race condition for limited I/O resource, the services or processes with lower priority occasionally occupy most of limited I/O resource. And it decreases QoS and performance of important services and makes it difficult to efficiently use limited I/O resource. Although system administrator allocates I/O resource according to priority of process, he/she can't know or expect how much resource will be used by the specific process. Due to these reasons, it causes the problem that he/she can't guarantee the service QoS and performance stability. Therefore, in this paper, we propose service based disk I/O control supporting predictable I/O bandwidth to resolve upper problems. Proposed I/O control guarantees the service QoS and performance stability by supporting the service based predictable I/O bandwidth and it makes limited I/O resource to be efficiently used in respect of service.

• Journal of KIISE
• /
• v.44 no.5
• /
• pp.469-475
• /
• 2017

The emerging NVMe-based multi-queue SSDs provides a high bandwidth by parallel I/O, i.e., each core performs I/O through its dedicated queue in parallel with other cores. To provide a bandwidth share for each application with I/O, a fair-share scheduler that provides a bandwidth share to each core is required. In this study, we proposed a multi-core scalable fair-queuing algorithm for multi-queue SSDs. The algorithm adopts randomization to minimize the inter-core synchronization overheads and provides a weight-proportional bandwidth share to each core. The results of our experiments indicated that the proposed algorithm gives accurate bandwidth partitioning and outperforms the existing FlashFQ scheduler, regardless of the number of cores for a Linux kernel with block-mq.

• Journal of Industrial Technology
• /
• v.40 no.1
• /
• pp.19-23
• /
• 2020

In this paper, we proposed a PC-based CNC control system using EtherCAT-based servo drive and I/O device. The default communication of LinuxCNC is a parallel port, and data processing with high bandwidth is impossible. However, it is possible to apply various bandwidth devices through the application of EtherCAT, one of the industrial Ethernet communications with high bandwidth. Therefore, the hardware control method of LinuxCNC was applied through EtherCAT communication from the existing parallel port. Finally, through HAL configuration, I/O device operation check and 3-axis motion control proved the LinuxCNC system with EtherCAT.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.523-532
• /
• 2001

A Critical issue in the performance of a video-on-demand system is the required I/O bandwidth of the Video server in order to satisfy clients requests, and it is the crucial resource that may cause delay increasingly. Several approaches such as batching and piggybacking are used to reduce the I/O demand on the video server through sharing. Bathing approach is to make single I/O request for storage server by grouping the requests for the same object. Piggybacking is th policy for altering display rates of requests in progress for the same object to merge their corresponding I/O streams into a single stream, and serve it as a group of merged requests. In this paper, we propose a reservation-based hybrid disk bandwidth reduction policy that dynamically reserves the I/O stream capacity of a video server for popular videos according to the loads of video server in order to schedule the requests for popular videos immediately. The performance of the proposed policy is evaluated through simulations, and is compared with that of bathing and piggybacking. As a result, we know that the reservation-based hybrid disk bandwidth reduction policy provides better probability of service, average waithing time and percentage of saving in frames than batching and piggybacking policy.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.11
• /
• pp.1318-1332
• /
• 1999

본 논문에서는 입출력 응용을 위해 collective I/O 기법을 기반으로 한 실행시간 시스템의 설계, 구현 그리고 그 성능평가를 기술한다. 여기서는 모든 프로세서가 동시에 I/O 요구에 따라 스케쥴링하며 I/O를 수행하는 collective I/O 방안과 프로세서들이 여러 그룹으로 묶이어, 다음 그룹이 데이터를 재배열하는 통신을 수행하는 동안 오직 한 그룹만이 동시에 I/O를 수행하는 pipelined collective I/O 등의 두 가지 설계방안을 살펴본다. Pipelined collective I/O의 전체 과정은 I/O 노드 충돌을 동적으로 줄이기 위해 파이프라인된다. 이상의 설계 부분에서는 동적으로 충돌 관리를 위한 지원을 제공한다. 본 논문에서는 다른 노드의 메모리 영역에 이미 존재하는 데이터를 재 사용하여 I/O 비용을 줄이기 위해 collective I/O 방안에서의 소프트웨어 캐슁 방안과 두 가지 모형에서의 chunking과 온라인 압축방안을 기술한다. 그리고 이상에서 기술한 방안들이 입출력을 위해 높은 성능을 보임을 기술하는데, 이 성능결과는 Intel Paragon과 ASCI/Red teraflops 기계 상에서 실험한 것이다. 그 결과 응용 레벨에서의 bandwidth는 peak point가 55%까지 측정되었다.Abstract In this paper we present the design, implementation and evaluation of a runtime system based on collective I/O techniques for irregular applications. We present two designs, namely, "Collective I/O" and "Pipelined Collective I/O". In the first scheme, all processors participate in the I/O simultaneously, making scheduling of I/O requests simpler but creating a possibility of contention at the I/O nodes. In the second approach, processors are grouped into several groups, so that only one group performs I/O simultaneously, while the next group performs communication to rearrange data, and this entire process is pipelined to reduce I/O node contention dynamically. In other words, the design provides support for dynamic contention management. Then we present a software caching method using collective I/O to reduce I/O cost by reusing data already present in the memory of other nodes. Finally, chunking and on-line compression mechanisms are included in both models. We demonstrate that we can obtain significantly high-performance for I/O above what has been possible so far. The performance results are presented on an Intel Paragon and on the ASCI/Red teraflops machine. Application level I/O bandwidth up to 55% of the peak is observed.he peak is observed.

• Annual Conference of KIPS
• /
• 2000.10a
• /
• pp.657-660
• /
• 2000

병렬 파일 시스템은 늦은 디스크 I/O로 인한 성능 저하를 개선하기 위해 병렬 I/O를 제공한다. 이때 계산과 디스크 I/O를 중첩시키는 선반입 기법으로 디스크 I/O로 인한 성능 저하를 더욱 개선할 수 있다. 하지만 I/O 위주의 프로그램에서는 선반입으로 인하여 시스템에서 제공하는 I/O 대역폭을 넘어 최악의 경우 기존의 선반입 기법은 성능개선을 위한 최선이 될 수 없을 뿐 아니라 선반입 기법 자체가 과부하가 될 수 있다. 본 논문에서는 이런 상황을 고려하여 I/O 대역폭 개선을 위한 이단 선반입 기법을 제시하여 성능개선을 제공한다.

• Journal of KIISE
• /
• v.42 no.11
• /
• pp.1322-1331
• /
• 2015

According to the hypervisor scheduler, the vCPU (virtual CPU) operates under two states: the running state and the stop state. When the vCPU is in the stop state, incoming events are delayed until that vCPU's state changes to the running state. The latency in handling such events that are sent to the vCPU is regarded as the I/O latency. Since a SMP (symmetric multiprocessing) VM (virtual machine) incorporates multiple vCPUs, the event latency on a SMP VM can vary according to specific vCPU that receives the event. In this paper, we propose a new scheme named event routing that sends events according to the operation state of each vCPU to reduce the event latency on an SMP VM. We implemented the proposed event routing scheme in Xen ARM hypervisor and confirmed the reduction of I/O latency from measuring the network RTT (round trip time) and the TCP bandwidth under a variety of testing conditions. The network RTT decreases by up to 94% and the TCP bandwidth increases up to 35% when compare to native Xen ARM.