BOSS: An Efficient Data Distribution Strategy for Object Storage Systems With Hybrid Devices

Wu, Liqun; Zhuge, Qingfeng; Sha, Edwin Hsing-Mean; Chen, Xianzhang; Cheng, Linfeng

doi:10.1109/access.2017.2744259

Cited by 13 publications

(7 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the read or write load is high, the delay of OSD in responding to the client's read and write data request will increase, leading to the death of the OSD node in extreme cases. (3) The number of OSDs on heterogeneous nodes is H, and the OSD type is T. The literature [29] presented a significant performance difference between SSD and HDD type OSDs in handling client read and write operations. (4) The number of PGs occupied by the OSD is P. Two implementations of the Object-Store, FileStore, and BlueStore, provide APIs for reading and writing threads, both at a PG granularity.…”

Section: Osd Performance Impact Factorsmentioning

confidence: 99%

An adaptive read/write optimized algorithm for Ceph heterogeneous systems via performance prediction and multi-attribute decision making

Wang

2022

Cluster Comput

View full text Add to dashboard Cite

The advent of the Big Data era has brought considerable challenges to storing and managing massive data. Moreover, distributed storage systems are critical to the pressure and storage capacity costs. The Ceph cloud storage system only selects data storage nodes based on node storage capacity. This node selection method results in load imbalance and limited storage scenarios in heterogeneous storage systems. Therefore, we add node heterogeneity, network state, and node load as performance weights to the CRUSH algorithm and optimize the performance of the Ceph system by improving load balancing. We designed a cloud storage system model based on Software Defined Network (SDN) technology. This system model can avoid the tedious configuration and significant measurement overhead required to obtain network status in traditional network architecture. Then we propose adaptive read and write optimization algorithms based on SDN technology. The Object Storage Device (OSD) is initially classified based on the Node Heterogeneous Resource Classification Strategy. Then the SDN technology is used to obtain network and load conditions in real-time and an OSD performance prediction model is built to obtain weights for performance impact factors. Finally, a mathematical model is proposed for multi-attribute decision making in conjunction with the OSD state and its prediction model. Furthermore, this model is addressed to optimize read and write performance adaptively. Compared with the original Ceph system, TOPSIS_PA improves the performance of reading operations by 36%; TOPSIS_CW and TOPSIS_PACW algorithms improve the elastic read performance by 23 to 60% and 36 to 85%, and the elastic write performance by 180 to 468% and 188 to 611%, respectively.

show abstract

Section: Osd Performance Impact Factorsmentioning

confidence: 99%

An adaptive read/write optimized algorithm for Ceph heterogeneous systems via performance prediction and multi-attribute decision making

Wang

2022

Cluster Comput

View full text Add to dashboard Cite

show abstract

“…There has also been work that looks at the stimulus responses of a disk [159] and models the dynamic power characteristics for a historic workload IO traces [160], which in turn is used for predicting the energy characteristics and optimization for energy efficiencies [161]. The approach with [168] assesses the data patterns and distributes the data onto a hybrid set of devices. These are also followed inside enterprise-class storage arrays which are hosted in Data Center, by implementing heuristics-based policies to drive the data into a heterogeneous set of disks like SSD and HDDs both through initial allocation and through automatic migration [169].…”

Section:  Energy Efficienciesmentioning

confidence: 99%

Optimizing Energy Efficiencies in Cloud Data Center Resources with Availability Constraints

Phalachandra,

Sitaram

2020

IJEAT

View full text Add to dashboard Cite

Cloud infrastructure Resources hosted in Data Centers, support the effective execution of Cloud computing applications. Given the increased adoption of the Cloud Computing Applications and the Businesses getting to be Data-driven, there is a huge increase in the number of Data Centers and the Size and amount of resources hosted in these Data Centers. These Data Center resources consume a significant amount of energy and this continuous scaling of the resources is leading to increased power consumption and a large carbon footprint. Given our fragile eco-system, optimization of the Data Center resources for energy conservation and thus the carbon footprint is the primary area of our focus. Businesses also need to satisfy QoS guarantees on Availability to their customers. Optimization towards Energy efficiencies may compromise on the Availability and thus may warrant a trade-off, and a need for them to be considered together. Although there have been numerous studies towards Energy efficiencies, most of them have been focused on only energy. In this paper, we initially segregate Optimization activities towards the Data Center resources like Compute, Network, and Storage. We then study the different control parameters or approaches which will lead to meeting the objectives of Energy Efficiencies, Availability and Energy Efficiency constrained with Availability. Thus, this will support the selection of approaches for the optimization of energy while meeting the QoS Availability requirement.

show abstract

“…Graphene [28] suggests a DRAM-SSD architecture to improve performance of graph computing for large graphs. SSD caching is also suggested in distributed and High Performance Computing (HPC) environments [29], [30], [31], [32], [33].…”

Section: Previous Studiesmentioning

confidence: 99%

An Efficient Hybrid I/O Caching Architecture Using Heterogeneous SSDs

Salkhordeh

Hadizadeh

Asadi

2019

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Storage subsystem is considered as the performance bottleneck of computer systems in data-intensive applications. Solid-State Drives (SSDs) are emerging storage devices which unlike Hard Disk Drives (HDDs), do not have mechanical parts and therefore, have superior performance compared to HDDs. Due to the high cost of SSDs, entirely replacing HDDs with SSDs is not economically justified. Additionally, SSDs can endure a limited number of writes before failing. To mitigate the shortcomings of SSDs while taking advantage of their high performance, SSD caching is practiced in both academia and industry. Previously proposed caching architectures have only focused on either performance or endurance and neglected to address both parameters in suggested architectures. Moreover, the cost, reliability, and power consumption of such architectures is not evaluated. This paper proposes a hybrid I/O caching architecture that while offers higher performance than previous studies, it also improves power consumption with a similar budget. The proposed architecture uses DRAM, Read-Optimized SSD (RO-SSD), and Write-Optimized SSD (WO-SSD) in a three-level cache hierarchy and tries to efficiently redirect read requests to either DRAM or RO-SSD while sending writes to WO-SSD. To provide high reliability, dirty pages are written to at least two devices which removes any single point of failure. The power consumption is also managed by reducing the number of accesses issued to SSDs. The proposed architecture reconfigures itself between performance-and enduranceoptimized policies based on the workload characteristics to maintain an effective tradeoff between performance and endurance. We have implemented the proposed architecture on a server equipped with industrial SSDs and HDDs. The experimental results show that as compared to state-of-the-art studies, the proposed architecture improves performance and power consumption by an average of 8% and 28%, respectively, and reduces the cost by 5% while increasing the endurance cost by 4.7% and negligible reliability penalty.

show abstract

BOSS: An Efficient Data Distribution Strategy for Object Storage Systems With Hybrid Devices

Cited by 13 publications

References 40 publications

An adaptive read/write optimized algorithm for Ceph heterogeneous systems via performance prediction and multi-attribute decision making

An adaptive read/write optimized algorithm for Ceph heterogeneous systems via performance prediction and multi-attribute decision making

Optimizing Energy Efficiencies in Cloud Data Center Resources with Availability Constraints

An Efficient Hybrid I/O Caching Architecture Using Heterogeneous SSDs

Contact Info

Product

Resources

About