Block Storage Research Articles

Experimental Study on Block and Ratio Storage Strategy and Parallel Transmission Algorithms

Parallel transmission algorithms CLBA and DAS greatly improved the speed of transferring data files, but they cause a large overhead of storage space and network traffic for deployment of duplications of large files, in addition to their poor feasibility of deploying large complete replicas on nodes with limited storage space. In view of Google and Hadoop distributed file systems all adopted block storage mechanism, a parallel data transmission algorithm based on block storage strategy is proposed, but the interrelation between block storage and parallel transmission was not properly taken into consideration and so the nodes of slower bandwidth greatly affected the overall transmission performance. Therefore, we proposed in this paper block and ratio storage strategies for files of massive data in distributed systems and parallel transmission algorithms PTBM and PTRM for the two storage strategies. Experimental results indicate the proposed algorithms are better than CLBA, and close to DAS, and they save more than 50% storage space, and can adapt to the deployment of large data files.

Modeling of PV Standalone Microgrid Based on IEEE Standards, 1562-2007, 1361-2003 and 1013-2007

Rooftop solar panels and solar farms has become popular as a means of generating green and emission free electric energy. A standalone Microgrid is fully controlled by sophisticated controlling scheme and ready to contribute to utility grid. A standalone solar Microgrid is an economic solution in residential area as well in far off location. It consists of PV panels, DC/DC converters, storage blocks, MPPT controller, Inverter, main controller and loads. A standalone PV system has two stages design. In power stage, all components will properly be sized to supply demand loads. Control stage contains all controllers to meet the predefined requirements. There is plenty of research on each individual element; however, for system integrator there are only a few studies which have considered the whole system. This study tries to show a systematic step design for Photovoltaic-based (PV) Microgrid. All power and control design stages will be discussed. IEEE standards and recommendation for standalone PV grid and storage banks (batteries) are used as guideline.

마모도 평준화를 위한 File Clustering 알고리즘

Storage device based on Flash Memory have many attractive features such as high performance, low power consumption, shock resistance, and low weight, so they replace HDDs to a certain extent. An Storage device based on Flash Memory has FTL(Flash Translation Layer) which emulate block storage devices like HDDs. A garbage collection, one of major functions of FTL, effects highly on the performance and the lifetime of devices. However, there is no de facto standard for new garbage collection algorithms. To solve this problem, we propose File Clustering Algorithm. File Clustering Algorithm respect to update page from same file at the same time. So, these are clustered to same block. For this mechanism, We propose Page Allocation Policy in FTL and use MIN-MAX GAP to guarantee wear leveling. To verify the algorithm in this paper, we use TPC Benchmark. So, The performance evaluation reveals that the proposed algorithm has comparable result with the existing algorithms(No wear leveling, Hot/Cold) and shows approximately 690% improvement in terms of the wear leveling.

Multiple Yard Cranes Scheduling Model and Algorithm in the Mixture Storage Block

In order to improve the operating efficiency of mixture storage yard, the problem of multiple yard cranes scheduling (MYCS) must to be solved. The paper develops a multiple yard cranes scheduling MIP model in order to minimize the sum of trucks’ waiting cost and yard cranes’ moving cost. An important feature is that the future has to be taken into account: the interference between two yard cranes, and the priorities of internal and external trucks, and the different arrival time of trucks and other factors. The study proposes an improved genetic algorithm (GA) to solve the MYCS problem and verifies the performance of algorithm by two numerical examples. The experimental results indicate that the proposed scheduling model is superior to the traditional scheduling and the algorithm is practically meaningful.

Research and Application of Location Assignment and Routing Strategies in Block Storage Yard of Shipbuilding

In shipbuilding, the storage yard is used to store assembly blocks of ships. It is an essential part of the shipbuilding system.The storage yard management contains:determining the exact storage locations for the arriving blocks, and determining the storing or retrieving route for each block. Aheuristic approach was proposedto maximize the space utilization and minimize the obstructive. Based on the approach, a storage yard management systemwas built.The system is used by the planners and got positive effect.

Problems of informational support of chemical and radiation ecological monitoring of long-term technological storage of reactor blocks of utilized submarines afloat in the northwestern region

Complexity of the problems related to utilization of nuclear submarines (NSM) is determined not only by their huge scale but also by the necessity of systemic consideration of diverse factors, including political, economic, ecological, social, technological, and informational. Afloat storage of reactor compartment units is a complex chemical, technological and ecological problem, for sea water induces their extensive corrosion which creates a risk of leakage of radioactive heavy metal isotopes, fission products, and transuranium elements to the environment. While searching for and testing of technological solutions for utilization of NSM, the most important problems are chemical and radiation ecological monitoring of water areas and its informational support. The article contains systemic analysis of problems related to long-term technological storage of reactor compartment units afloat, and an optimal procedure for ecological radionuclide monitoring is proposed. The proposed procedure requires minimal equipment and radiation control range but ensures a sufficient confidence level.

The Effect of Porous Block Pavement with Rainwater Storage on Heat Island Mitigation

The recent abnormal climate and extreme weather events have frequently given unexpected casualties and damages. In particular, in cases accompanying with heavy rainfall or extremely hot weather in a short period of time, there increases possibility leading financial damage by flooding, heat island phenomenon etc.. One of the main factors that are caused these problems is impermeable area including asphalt and concrete pavement which is increased by urbanization. So, it is interested in porous pavement to solve the environmental problems in Korea. In this study, a block pavement system for sidewalk to control surface temperature of pavement and rainwater runoff was developed. The block pavement system is composed of permeable or water-retentive block and rainwater storage which can harvest and supply rainwater. Surface temperature of permeable block pavement is reduced about 10°C compare with it of impermeable block pavement. Also, rainwater runoff was not happened during the period of rainfall.

플래시 SSD의 내구성을 검증하기 위한 FTL 오프라인 알고리즘

SSDs(Solid State Drives) have many attractive features such as high performance, low power consumption, shock resistance, and low weight, so they replace HDDs to a certain extent. An SSD has FTL(Flash Translation Layer) which emulate block storage devices like HDDs. A garbage collection, one of major functions of FTL, effects highly on the performance and the lifetime of SSDs. However, there is no de facto standard for new garbage collection algorithms. To solve this problem, we propose trace driven offline optimal algorithms for garbage collection of FTL. The proposed algorithm always guarantees minimal number of erase operation. In addition, we verify our proposed algorithm using TPC trace.

Block storage virtualization with commodity secure digital cards

Smartphones, tablets and other mobile platforms typically accommodate bulk data storage with low-cost, FAT-formatted Secure Digital cards. When one uses a mobile device to run a full-system virtual machine (VM), there can be a mismatch between 1) the VM's I/O mixture, security and reliability requirements and 2) the properties of the storage media available for VM block storage and checkpoint images. To resolve this mismatch, this paper presents a new VM disk image format called the Logging Block Store (LBS). After motivating the need for a new format, LBS is described in detail with experimental results demonstrating its efficacy. As a result of this work, recommendations are made for future optimizations throughout the stack that may simplify and improve the performance of storage virtualization systems on mobile platforms.

From paxos to CORFU

Introduction In a 2006 talk [11], Jim Gray predicted that “flash is the new disk”; five years later, indeed, flash is making headway into data centers, but it is usually disguised as standard block storage. We posit that new abstractions are required to exploit the potential of large-scale flash clusters. It has been observed that each individual flash drive is best utilized in a log-structured manner due to the intrinsic properties of flash [6]. Further to that, in order to harness the aggregate bandwidth of a cluster of flash units, we propose to organize the entire cluster as a single shared log. CORFU is a new storage system for flash clusters which demonstrates the feasibility of this approach. The key idea in CORFU is to expose a cluster of network-attached flash devices as a single, shared log to clients running within the data center. Applications running on the clients can append data to this log or read entries from its middle.

Sequencing heuristics for storing and retrieving unit loads in 3D compact automated warehousing systems

Sequencing unit-load retrieval requests has been extensively reported on in the literature for conventional single-deep automated warehousing systems. A proper sequence can greatly reduce the makespan when carrying out a group of such requests. Although the sequencing problem is NP-hard, some very good heuristics exist. Surprisingly, the problem has not yet been investigated for compact (multi-deep) storage systems, which have greatly increased in popularity the last decade. This article studies how to sequence a group (or block) of storage and retrieval requests in a multi-deep automated storage system with the objective to minimize the makespan. Currently utilized sequencing heuristics for the multi-deep system are adapted in this article and in addition a new heuristic, Percentage Priority to Retrievals with Shortest Leg (PPR-SL), is proposed and evaluated. It is shown that the PPR-SL heuristic consistently outperforms all of the other heuristics. Generally, it can outperform the benchmark First-Come First-Served (FCFS) heuristic by between 20 and 70%. The nearest neighbor heuristic that performs very well in conventional single-deep storage systems appears to perform poorly in the multi-deep system, even worse than FCFS. In addition, based on FCFS and PPR-SL, robust rack dimensions that yield a short makespan, regardless of the number of storage and retrieval requests, are found.

Managing State

The notion of state is fundamental to the study of dynamical systems. State refers to the minimal set of information that fully describes the system at a specific time. The simplest state model is found in Fortran 77, where a program's state consists of the values of all its scalar variables and of all the elements of all arrays, the statement to be executed next, and the statements at which execution resumes at the end of each active subroutine or function. In the following, I'll concentrate on the variables, because the statement references are part of control flow and not directly accessible to the programmer. It's important, however, to remember that each state is associated with a precise point in the program's execution. The two important concepts in understanding data handling in Fortran 77 are thus value and variable, where the latter term includes the elements of arrays. Each variable has a value at each point in execution. This value is changed by an assignment statement. It can be used in an expression, where it's combined with other values (constants or the values of other variables) by operations, such as arithmetic, to produce new values. Languages such as C or Fortran 90 that offer pointers and dynamic memory allocation lead to a much more complicated description of state. Obviously the contents of the dynamically allocated storage must be included in the total state of a program, but that's only the first step. Each such storage block can itself contain pointers to other storage blocks. The neat distinction between variables and values disappears: the value of a variable can define another variable. As a consequence, it's no longer possible to compile a list of all parts of a program's state just by looking at its source code. What is or isn't part of the state is defined by the state itself.

컨테이너 픽업문제를 위한 유전자 알고리듬

Container pick-up scheduling problem is to minimize the total container handling time, which consists of the traveling distance and the setup time of yard cranes in a container yard. Yard cranes have to pick-up the containers which are stacked in the yard-bays to satisfy the work schedule requirement of quay crane, which loads and unloads containers on or from container ships. This paper allows the movement of multiple yard cranes among storage blocks. A mixed integer programming model has been formulated and a genetic algorithm (GA) has been proposed to solve problems of large sizes. Computational results show that the proposed GA is an effective method.

A comparison of immunohistochemistry and in situ hybridization for the detection of porcine circovirus type 2 in pigs

The aim of this study was to develop and to optimize an immunohistochemistry (IHC) method for PCV2 identification and to compare it with an in situ hybridization (ISH) technique. The results demonstrated that both ISH and IHC successfully detected PCV2 viral antigens or nucleic acid in the examined tissues. Most of the slides identified previously in ISH as PCV2-positive were also positive in IHC. In the case of nearly half of the slides the results of IHC examination revealed an increase in the intensity of staining. IHC presented higher sensitivity and specificity than ISH. No negative impact of the time of paraffin block storage on ISH detection results was observed. In addition, IHC results were easier to interpret due to better image quality after staining. Overall results confirmed IHC was a reliable and useful technique for PMWS diagnosis.

Effects of Preanalytical Variables on the Detection of Proteins by Immunohistochemistry in Formalin-Fixed, Paraffin-Embedded Tissue

While formalin fixation and paraffin embedding has become a universal mechanism of tissue preservation and a gold standard for immunohistochemistry, fixation and processing variables that may confound assay effectiveness have received little attention from the scientific community. To identify discrete steps in specimen fixation and processing that may impact immunostaining, assess the magnitude of reported effects in the literature, and highlight preanalytical variables that require further investigation. Thirty-nine primary research articles that investigated immunohistochemical effects of 1 or more preanalytical variables were identified by our literature survey. Thresholds identified in the literature were then compared with published immunohistochemistry guidelines for formalin-fixed, paraffin-embedded specimens. Of the 62 preanalytical variables identified, 27 were examined in published research. Meta-analysis revealed 15 preanalytical variables that were capable of impacting immunohistochemistry (including fixation delay; fixative type; time in fixative; reagents and conditions of dehydration, clearing, and paraffin impregnation; and conditions of slide drying and storage) and 12 variables with no reported influence (including the type of processor used; the number and position of specimens during dehydration, clearing, and paraffin impregnation; and the duration of paraffin block storage). Variables with antigen-dependent or inconsistent effects were highlighted. Comparison of literature-supported thresholds with published recommendations revealed (1) strong agreement among preanalytical variables for optimal immunostaining, (2) discrepancies among thresholds for adequate immunostaining, and (3) the continued need for rigorous research and comprehensive guidelines on specimen fixation, processing, and storage.

Water flow and storage in fractured, unsaturated sulphur blocks

Water is a primary control on the generation of acidic (H2SO4) effluent from commercial-scale sulphur (S0) blocks. Although clean S0 is strongly hydrophobic, microbial colonization of fracture faces and friable S0 generates localized hydrophilic conditions. Infiltration occurs preferentially along discrete fractures and in areas of friable S0. Surface evaporation rates are low (mean 0.2 mm/day), and >90% of rainfall infiltrates and rapidly drains from the blocks. A conceptual model to describe the flow and storage of water at the base of S0 blocks was developed and tested. The S0 blocks were represented as a hydrophilic equivalent porous medium, and lateral drainage through the base of the blocks was quantified using a two-dimensional numerical model. Specific yield (Sy) and saturated hydraulic conductivity (Ks) values were estimated to be ∼4%–7% and ∼1 × 10−2 m/s, respectively, by comparing measured and modelled pore-water pressure head and outflow responses to rainfall events. Given that commercial-scale S0 blocks are constructed in a similar manner worldwide, the results of this study are considered widely applicable in designing S0 block storage facilities that minimize water availability and H2SO4 production in S0 blocks.

Utilizing mRNA extracted from small, archival formalin-fixed paraffin-embedded prostate samples for translational research: assessment of the effect of increasing sample age and storage temperature

Translational prostate cancer research is hampered by long intervals from diagnosis to patient progression and difficulty in obtaining cancer tissue for investigation. As such, it is imperative to utilise aging formalin-fixed paraffin-embedded (FFPE) tissue samples from the pathology archive with linked patient outcome data to allow current day research. This study aimed to assess the adequacy and quantity of mRNA extracted from formalin fixed paraffin embedded (FFPE) prostate tissue, including prostate biopsies, up to 15 years old. The decay of mRNA over time and under differing storage conditions was also assessed. Archived FFPE benign prostatic tissue up to 15 years old from transurethral resection of the prostate (TURP) and transrectal ultrasound guided (TRUS) biopsies as well as fresh tissue obtained from patients undergoing TURP for benign bladder outlet obstruction were used. Following mRNA extraction beta-actin real-time PCR was carried out using a set of 4 different primer/probes to assess mRNA quality and quantity. There was no difference in mRNA quantity/quality extracted from "fresh" FFPE tissue from the same patient over a 4-month period following surgery. The temperature of block storage did not alter quality/quantity of the mRNA (P > 0.05, unpaired t test). Fresh tissue had a higher quality/quantity, indicated by a lower C ( T ) value, than FFPE samples from the same patient (P ≤ 0.03, one-way ANOVA). Despite being up to 15 years old, all archived FFPE TURP and TRUS biopsy samples had "high" or "very high" levels of expression making them suitable for further analysis. However, the quality of the mRNA in archived FFPE samples did significantly decline with increasing sample age. It is possible to extract mRNA of sufficient standard for further transcriptomic analysis from minute FFPE samples up to 15 years old. This work adds to the literature suggesting that exploitation of retrospective prostate tissue collections with robust associated clinical data is possible.

Effects of storage block layout and automated yard crane systems on the performance of seaport container terminals

As a decoupling point between waterside and landside transport, the container yard plays a major role for the competitiveness of container terminals. One of the latest trends in container yard operations is the automated rail-mounted-gantry-crane system, which offers dense stacking along with high productivity. In this paper, the strategic design of rail-mounted-gantry-crane systems is investigated. A simulation study is conducted to evaluate the effects of four rail-mounted-gantry-crane systems and 385 yard block layouts--differing in block length, width, and height--on the yard and terminal performance.

An Improved Algorithm About Failure Agreement Based on Distributed Shared Memory

According to the uniform addressing and direct localization of network address space, adopting multi-threaded, multi-copy store and block storage in pages, page tree sub-node parallel failure methods, these can improve the efficiency and parallel processing of the DSM system. It is significance for improving the performance of distributed systems.

Virtual machine file system

The Virtual Machine File System (VMFS) is a scalable and high performance symmetric clustered file system for hosting virtual machines (VMs) on shared block storage. It implements a clustered locking protocol exclusively using storage links, and does not require network-based inter-node communication between hosts participating in a VMFS cluster. VMFS layout and IO algorithms are optimized towards providing raw device speed IO throughput to VMs. An adaptive IO mechanism masks errors on the physical fabric using contextual information from the fabric. The VMFS lock service forms the basis of VMware's clustered applications such as vMotion, Storage vMotion, Distributed Resource Scheduling, High Availability, and Fault Tolerance. Virtual machine metadata is serialized to files and VMFS provides a POSIX interface for cluster-safe virtual machine management operations. It also contains a pipelined data mover for bulk data initialization and movement. In recent years, VMFS has inspired changes to diskarray firmware and the SCSI protocol. These changes enable the file system to implement a hardware accelerated data mover and lock manager, among other things. In this paper, we present the VMFS architecture and its evolution over the years