A high-availability software update method for distributed storage systems

Kobayashi, D.; Watanabe, Akitsugu; Uehara, Tetsutaro; Yokota, Haruo

doi:10.1002/scj.20503

Cited by 3 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Storage systems need to provide high IO transactions in addition to their capacity and availability requirements [7,9]. Generally large and medium size storage systems consist of disk arrays and associated caches to improve the IO performance.…”

Section: High Io Intensive Information System Architecturementioning

confidence: 99%

Layered Memory Architecture for High IO Intensive Information Services to Achieve Timeliness

Takahashi

Ahmad

Möri

2008

2008 11th IEEE High Assurance Systems Engineering Symposium

View full text Add to dashboard Cite

Storage systems need to provide high IO transactions in addition to their capacity and availability requirements to achieve high assurance characteristics, such as fault tolerance and timeliness. Caches built in large and medium size storage systems have not been very effective to improve the IO transactions and overall performance of the system. Caches at various layers of the memory hierarchy in computing systems are more fast but very small compared to the adjacent lower level of storage device. This paper proposes a novel concept to exploit local memory as a block device to improve the IO performance of storage systems called Data Transmission System (DTS) concept. This concept is based on hierarchical layered memory utilization and management in the information system. Data accessed frequently is maintained on the managed DTS cache device in close proximity to the CPU. The proposed concept is substantially different from the conventional unmanaged cache in the computing systems. System architecture and detailed system design are presented in this paper to realize the proposed concept. Performance evaluation of the system has been carried out through numerical as well as application scenario. The experimental results show improvement of the order of 10 to 100 times faster IO transactions and reduce CPU wait time significantly.Keyword: DTS (Data Transmission System) cache, Timeliness, Write back cache, Block IO device, Locality of reference, IOPS, Bonnie Bench mark program, iSCSI protocol. Back groundThere are numerous Web based applications and information services available today. Some of the applications can be characterized as high IO (Input Output) intensive such as video on demand, video services, and medical sciences including healthcare and digital imaging. A number of application domains such as data mining need high IO intensive system to access data from/to hard disk as it cannot fit into the main memory.Consequently, Information technology is enhancing rapidly and it is bringing more useful, convenient, and advance human society system. These emerging applications require continuous operation, non stop service system and timeliness to achieve high assurance [1,3]. However, realization of such system requires the characteristics of Autonomous Decentralized System (ADS) [2,4]. Theoretical foundations of ADS rely on the principles of autonomous controllability and autonomous coordinatability. These two properties assure timeliness, online expansion, fault tolerance and online maintenance of the system. Moreover, the information system computing facility in data centers consume high energy and electric power, mostly for its storage systems. Therefore these systems emit huge amount of CO2 by heat dissipation. There is need to consider how to design information system to achieve efficiency for power consumption. To reduce the electric power consumption on information system, the paper focuses on system characteristics for IO requests. What are the factors of bottleneck and from where these arise? General...

show abstract

Section: High Io Intensive Information System Architecturementioning

confidence: 99%

Layered Memory Architecture for High IO Intensive Information Services to Achieve Timeliness

Takahashi

Ahmad

Möri

2008

2008 11th IEEE High Assurance Systems Engineering Symposium

View full text Add to dashboard Cite

show abstract

“…The concept of Ramdisk for Linux/windows is quite prevalent but it is simply virtual drive on top of local memory and does not involve any cache management technique [11]. In contrast, the proposed platinum cache is a block-layer cache and any application can use it [12]. To gain performance, applications just need to be installed on top of L 3 cache.…”

Section: Related Workmentioning

confidence: 99%

Autonomous L3 Cache Technolgy for High Responsiveness

Takahashi

Malik

Möri

2012

Journal of Information Processing

View full text Add to dashboard Cite

Web services and cloud computing paradigms have opened up many new vistas. The data intensive cloud applications usually require huge amounts of data to input and output from secondary storage systems. The outstanding progress in area of network communications has enabled high speed networks and therefore, communication latency bottleneck in cloud and other web applications has been shifted to node/storage level. Moreover, existing cloud solutions focused mainly on the efficient utilization of computing resources through virtualization and issues of storage bottleneck did not receive much attention. Moreover, virtualization based implementation ensures equal priority to all hosted applications, thus, real time applications in cloud environment can't meet their requirements. To meet the demand of overall low latency in cloud and other web services; and particularly to reduce I/O bottleneck at storage level, novel idea of autonomous L3 cache technology is proposed. Autonomous L3 cache technology utilizes local memory space as dedicated block device cache for certain specific application, thus prioritizing it over rest of hosted ones. Evaluation shows performance improvement of 5-8 times in terms of timeliness in given setup.

show abstract