Distributed File System Virtualization Techniques Supporting On-Demand Virtual Machine Environments for Grid Computing

This paper presents a data management solution which allows fast Virtual Machine (VM) instantiation and efficient run-time execution to support VMs as execution environments in Grid computing. It is based on novel distributed file system virtualization techniques and is unique in that: (1) it provides on-demand cross-domain access to VM state for unmodified VM monitors; (2) it enables private file system channels for VM instantiation by secure tunneling and session-key based authentication; (3) it supports user-level and write-back disk caches, per-application caching policies and middleware-driven consistency models; and (4) it leverages application-specific meta-data associated with files to expedite data transfers. The paper reports on its performance in wide-area setups using VMware-based VMs. Results show that the solution delivers performance over 30% better than native NFS and with warm caches it can bring the application-perceived overheads below 10% compared to a local-disk setup. The solution also allows a VM with 1.6 GB virtual disk and 320 MB virtual memory to be cloned within 160 seconds for the first clone and within 25 seconds for subsequent clones. Ming Zhao is a PhD candidate in the department of Electrical and Computer Engineering and a member of the Advance Computing and Information Systems Laboratory, at University of Florida. He received the degrees of BE and ME from Tsinghua University. His research interests are in the areas of computer architecture, operating systems and distributed computing. Jian Zhang is a PhD student in the Department of Electrical and Computer Engineering at University of Florida and a member of the Advance Computing and Information Systems Laboratory (ACIS). Her research interest is in virtual machines and Grid computing. She is a member of the IEEE and the ACM. Renato J. Figueiredo received the B.S. and M.S. degrees in Electrical Engineering from the Universidade de Campinas in 1994 and 1995, respectively, and the Ph.D. degree in Electrical and Computer Engineering from Purdue University in 2001. From 2001 until 2002 he was on the faculty of the School of Electrical and Computer Engineering of Northwestern University at Evanston, Illinois. In 2002 he joined the Department of Electrical and Computer Engineering of the University of Florida as an Assistant Professor. His research interests are in the areas of computer architecture, operating systems, and distributed systems.     

Failure-Atomic File Access in the Slice Interposed Network Storage System

This paper presents a recovery protocol for block I/O operations in Slice, a storage system architecture for high-speed LANs incorporating network-attached block storage. The goal of the Slice architecture is to provide a network file service with scalable bandwidth and capacity while preserving compatibility with off-the-shelf clients and file server appliances. The Slice prototype virtualizes the Network File System (NFS) protocol by interposing a request switching filter at the client's interface to the network storage system. The distributed Slice architecture separates functions typically combined in central file servers, introducing new challenges for failure atomicity. This paper presents a protocol for atomic file operations and recovery in the Slice architecture, and related support for reliable file storage using mirrored striping. Experimental results from the Slice prototype show that the protocol has low cost in the common case, allowing the system to deliver client file access bandwidths approaching gigabit-per-second network speeds.     

Fast VMM-based overlay networking for bridging the cloud and high performance computing

A collection of virtual machines (VMs) interconnected with an overlay network with a layer 2 abstraction has proven to be a powerful, unifying abstraction for adaptive distributed and parallel computing on loosely-coupled environments. It is now feasible to allow VMs hosting high performance computing (HPC) applications to seamlessly bridge distributed cloud resources and tightly-coupled supercomputing and cluster resources. However, to achieve the application performance that the tightly-coupled resources are capable of, it is important that the overlay network not introduce significant overhead relative to the native hardware, which is not the case for current user-level tools, including our own existing VNET/U system. In response, we describe the design, implementation, and evaluation of a virtual networking system that has negligible latency and bandwidth overheads in 1–10 Gbps networks. Our system, VNET/P, is directly embedded into our publicly available Palacios virtual machine monitor (VMM). VNET/P achieves native performance on 1 Gbps Ethernet networks and very high performance on 10 Gbps Ethernet networks. The NAS benchmarks generally achieve over 95 % of their native performance on both 1 and 10 Gbps. We have further demonstrated that VNET/P can operate successfully over more specialized tightly-coupled networks, such as Infiniband and Cray Gemini. Our results suggest it is feasible to extend a software-based overlay network designed for computing at wide-area scales into tightly-coupled environments.     

ORTHRUS: a lightweighted block-level cloud storage system

Taking advantage of distributed storage technology and virtualization technology, cloud storage systems provide virtual machine clients customizable storage service. They can be divided into two types: distributed file system and block level storage system. There are two disadvantages in existing block level storage system: Firstly, Some of them are tightly coupled with their cloud computing environments. As a result, it’s hard to extend them to support other cloud computing platforms; Secondly, The bottleneck of volume server seriously affects the performance and reliability of the whole system. In this paper we present a lightweighted block-level storage system for clouds—ORTHRUS, based on virtualization technology. We first design the architecture with multiple volume servers and its workflows, which can improve system performance and avoid the problem. Secondly, we propose a Listen-Detect-Switch mechanism for ORTHRUS to deal with contingent volume servers’ failure. At last we design a strategy that dynamically balances load between multiple volume servers. We characterize machine capability and load quantity with black box model, and implement the dynamic load balance strategy which is based on genetic algorithm. Extensive experimental results show that the aggregated I/O throughputs of ORTHRUS are significantly improved (approximately two times of that in Orthrus), and both I/O throughputs and IOPS are also remarkably improved (about 1.8 and 1.2 times, respectively) by our dynamic load balance strategy.     

PUNCH: An architecture for Web-enabled wide-area network-computing

This paper addresses the architectural issues that arise in the design of a universally accessible wide-area network-computing system that is capable of making automatic cost/performance tradeoff decisions at run-time. The core system is designed around a three-level hierarchically distributed architecture, a choice driven by the dynamic, incremental, and distributed nature of the information associated with run-time cost/performance tradeoff decisions. Support for independent replication of each component in the hierarchy contributes to the overall scalability and reliability of the architecture. Meta-information is managed in a scalable manner by employing self-encoded resource identifiers that allow O(1) access to all managed information. Security and access control across administrative domains are provided by partitioning the infrastructure into independently-managed cells, and by giving administrators the ability to customize user-views directly at the location at which the request is processed. Demand-driven resource management is achieved by predicting the run-specific resource usage characteristics of tools via machine learning techniques. The concepts described in this paper are embodied in the Purdue University Network-Computing Hubs (PUNCH), a demand-based network-computing system that allows users to access and run unmodified tools via standard World Wide Web browsers. Tools do not have to be written in any particular language, and access to the source or object code is not required. The PUNCH infrastructure can be distributed in a manner that allows tools to be (user-transparently) executed wherever they reside. Currently, PUNCH contains over thirty tools developed by eight universities and four vendors, and serves more than five hundred users. During the past three years, PUNCH users have logged more than one million hits and have performed over seventy thousand simulations. This revised version was published online in July 2006 with corrections to the Cover Date.     

Scheduling parallel applications in distributed networks

Prophet is a run-time scheduling system designed to support the efficient execution of parallel applications written in the Mentat programming language (Grimshaw, 1993). Prior results demonstrated that SPMD applications could be scheduled automatically in an Ethernet-based local-area workstation network with good performance (Weissman and Grimshaw, 1994 and 1995). This paper describes our recent efforts to extend Prophet along several dimensions: improved overhead control, greater resource sharing, greater resource heterogeneity, wide-area scheduling, and new application types. We show that both SPMD and task parallel applications can be scheduled effectively in a shared heterogeneous LAN environment containing ethernet and ATM networks by exploiting the application structure and dynamic run-time information. This revised version was published online in July 2006 with corrections to the Cover Date.     

Secure outsourcing of modular exponentiations in cloud and cluster computing

Cloud computing and cluster computing are user-centric computing services. The shared software and hardware resources and information can be provided to the computers and other equipments according to the demands of users. A majority of services are deployed through outsourcing. Outsourcing computation allows resource-constrained clients to outsource their complex computation workloads to a powerful server which is rich of computation resources. Modular exponentiation is one of the most complex computations in public key based cryptographic schemes. It is useful to reduce the computation cost of the clients by using outsourcing computation. In this paper, we propose a novel outsourcing algorithm for modular exponentiation based on the new mathematical division under the setting of two non-colluding cloud servers. The base and the power of the outsourced data can be kept private and the efficiency is improved compared with former works.     

The MOSIX Direct File System Access Method for Supporting Scalable Cluster File Systems

MOSIX is a cluster management system that supports preemptive process migration. This paper presents the MOSIX Direct File System Access (DFSA), a provision that can improve the performance of cluster file systems by allowing a migrated process to directly access files in its current location. This capability, when combined with an appropriate file system, could substantially increase the I/O performance and reduce the network congestion by migrating an I/O intensive process to a file server rather than the traditional way of bringing the file's data to the process. DFSA is suitable for clusters that manage a pool of shared disks among multiple machines. With DFSA, it is possible to migrate parallel processes from a client node to file servers for parallel access to different files. Any consistent file system can be adjusted to work with DFSA. To test its performance, we developed the MOSIX File-System (MFS) which allows consistent parallel operations on different files. The paper describes DFSA and presents the performance of MFS with and without DFSA.     

Managing performance and power consumption tradeoff for multiple heterogeneous servers in cloud computing

There are typically multiple heterogeneous servers providing various services in cloud computing. High power consumption of these servers increases the cost of running a data center. Thus, there is a problem of reducing the power cost with tolerable performance degradation. In this paper, we optimize the performance and power consumption tradeoff for multiple heterogeneous servers. We consider the following problems: (1) optimal job scheduling with fixed service rates; (2) joint optimal service speed scaling and job scheduling. For problem (1), we present the Karush-Kuhn-Tucker (KKT) conditions and provide a closed-form solution. For problem (2), both continuous speed scaling and discrete speed scaling are considered. In discrete speed scaling, the feasible service rates are discrete and bounded. We formulate the problem as an MINLP problem and propose a distributed algorithm by online value iteration, which has lower complexity than a centralized algorithm. Our approach provides an analytical way to manage the tradeoff between performance and power consumption. The simulation results show the gain of using speed scaling, and also prove the effectiveness and efficiency of the proposed algorithms.     

DENS: data center energy-efficient network-aware scheduling

In modern data centers, energy consumption accounts for a considerably large slice of operational expenses. The existing work in data center energy optimization is focusing only on job distribution between computing servers based on workload or thermal profiles. This paper underlines the role of communication fabric in data center energy consumption and presents a scheduling approach that combines energy efficiency and network awareness, named DENS. The DENS methodology balances the energy consumption of a data center, individual job performance, and traffic demands. The proposed approach optimizes the tradeoff between job consolidation (to minimize the amount of computing servers) and distribution of traffic patterns (to avoid hotspots in the data center network).     

File and Object Replication in Data Grids

Data replication is a key issue in a Data Grid and can be managed in different ways and at different levels of granularity: for example, at the file level or object level. In the High Energy Physics community, Data Grids are being developed to support the distributed analysis of experimental data. We have produced a prototype data replication tool, the Grid Data Mirroring Package (GDMP) that is in production use in one physics experiment, with middleware provided by the Globus Toolkit used for authentication, data movement, and other purposes. We present here a new, enhanced GDMP architecture and prototype implementation that uses Globus Data Grid tools for efficient file replication. We also explain how this architecture can address object replication issues in an object-oriented database management system. File transfer over wide-area networks requires specific performance tuning in order to gain optimal data transfer rates. We present performance results obtained with GridFTP, an enhanced version of FTP, and discuss tuning parameters.     

Native flower strips increase visitation by non-bee insects to avocado flowers and promote yield

Pollination is an essential ecosystem service for pollinator-dependent crops and plant communities. Apis mellifera L. is by far the most commonly used species to obtain this service in agriculture. However, there is growing evidence of the importance for crop yields of the service provided by wild bees and non-bee insect pollinators. Establishing flower resources in agricultural landscapes is a management practice that can increase insect pollinator populations and improve crop yields. We established perennial native flower strips (NFS) in four avocado orchards in central Chile during autumn 2017. We monitored flower visitors and counted newly formed fruits in avocados near and far from NFS in spring 2019, to assess flower visitor groups, flower visitation rates and fruit numbers. Only A. mellifera visited avocado flowers within bees, whereas both the managed A. mellifera and wild bees were the main visitors to the NFS. NFS increased visitation rates to adjacent avocado of flies, and with a trend for the sum of all non-managed flower visitors (i.e. excluding A. mellifera). However, there were no differences in the rates of A. mellifera visitation to avocados near and far from NFS. Avocado fruit numbers were higher among avocados near NFS than among those farther away. This difference could be due to better pollination by the increased visits to avocado flowers by flies or other wild insects. Therefore, NFS could contribute to crop fruit number, as well as conservation of native flora, wild bees and non-bee pollinators on fruit farms in the “Central Chile” biodiversity hotspot.     

ACS: An adaptive communication system for heterogeneous wide-area ATM clusters

This paper presents an architecture, implementation, and performance evaluation of an adaptive message-passing system for a heterogeneous wide-area ATM cluster that we call the Adaptive Communication System (ACS). ACS uses multithreading to provide efficient techniques for overlapping computation and communication in wide-area computing. By separating control and data activities, ACS eliminates unnecessary control transfers over the data path. This optimizes the data path and improves the performance. ACS supports several different flow control algorithms, error control algorithms, and multicasting algorithms. Furthermore, ACS allows programmers to select at runtime the suitable communication schemes per-connection basis to meet the requirements of a given application. ACS provides three application communication interfaces: Socket Communication Interface (SCI), ATM Communication Interface (ACI), and High Performance Interface (HPI) to support various classes of applications. The SCI is provided mainly for applications that must be portable to many different computing platforms. The ACI provides services that are compatible with ATM connection oriented services where each connection can be configured to meet the Quality of Service (QOS) requirements of that connection. This allows programmers to fully utilize the benefits of the ATM network. The HPI supports applications that demand low-latency and high-throughput communication services. In this interface, ACS uses read/write trap routines to reduce latency and data transfer time, and to avoid using traditional communication protocols. We analyze and compare the performance of ACS with those of other message-passing systems such as p4, PVM, and MPI in terms of point-to-point, multicasting, and application performance. The benchmarking results show that ACS outperforms other message-passing systems and provides flexible communication services for various classes of applications. This revised version was published online in July 2006 with corrections to the Cover Date.     

Adaptive hybrid storage systems leveraging SSDs and HDDs in HPC cloud environments

Cloud computing should inherently support various types of data-intensive workloads with different storage access patterns. This makes a high-performance storage system in the Cloud an important component. Emerging flash device technologies such as solid state drives (SSDs) are a viable choice for building high performance computing (HPC) cloud storage systems to address more fine-grained data access patterns. However, the bit-per-dollar SSD price is still higher than the prices of HDDs. This study proposes an optimized progressive file layout (PFL) method to leverage the advantages of SSDs in a parallel file system such as Lustre so that small file I/O performance can be significantly improved. A PFL can dynamically adjust chunk sizes and stripe patterns according to various I/O traffics. Extensive experimental results show that this approach (i.e. building a hybrid storage system based on a combination of SSDs and HDDs) can actually achieve balanced throughput over mixed I/O workloads consisting of large and small file access patterns.     

Scalable Session Locking for a Distributed File System

File systems provide an interface for applications to obtain exclusive access to files, in which a process holds privileges to a file that cannot be preempted and restrict the capabilities of other processes. Local file systems do this by maintaining information about the privileges of current file sessions, and checking subsequent sessions for compatibility. Implementing exclusive access in this manner for distributed file systems degrades performance by requiring every new file session to be registered with a lock server that maintains global session state. We present two techniques for improving the performance of session management in the distributed environment. We introduce a distributed lock for managing file access, called a semi-preemptible lock, that allows clients to cache privileges. Under a semi-preemptible lock, a file system creates new sessions without messages to the lock manager. This improves performance by exploiting locality – the affinity of files to clients. We also present data structures and algorithms for the dynamic evaluation of locks that allow a distributed file system to efficiently manage arbitrarily complex locking. In this case, complex means that an object can be locked in a large number of unique modes. The combination of these techniques results in a distributed locking scheme that supports fine-grained concurrency control with low memory and message overhead and with the assurance that their locking system is correct and avoids unnecessary deadlocks.     

BioDownloader: bioinformatics downloads and updates in a few clicks

There are many ftp or http servers storing data required for biological research. While some download applications are available, there is no user-friendly download application with a graphical interface specifically designed and adapted to meet the requirements of bioinformatics. BioDownloader is a program for downloading and updating files from ftp and http servers. It is optimized to work robustly with large numbers of files. It allows the selective retrieval of only the required files (batch downloads, multiple file masks, ls-lR file parsing, recursive search, recent updates, etc.). BioDownloader has a built-in repository containing the settings for common bioinformatics file-synchronization needs, including the Protein Data Bank (PDB) and National Center for Biotechnology Information (NCBI) databases. It can post-process downloaded files, including archive extraction and file conversions. AVAILABILITY: The program can be installed from http://dunbrack.fccc.edu/BioDownloader. The software is freely available for both non-commercial and commercial users under the BSD license.     

Performance implications of non-uniform VCPU-PCPU mapping in virtualization environment

Virtualization technology promises to provide better isolation and consolidation in traditional servers. However, with VMM (virtual machine monitor) layer getting involved, virtualization system changes the architecture of traditional software stack, bringing about limitations in resource allocating. The non-uniform VCPU (virtual CPU)-PCPU (physical CPU) mapping, deriving from both the configuration or the deployment of virtual machines and the dynamic runtime feature of applications, causes the different percentage of processor allocation in the same physical machine,and the VCPUs mapped these PCPUs will gain asymmetric performance. The guest OS, however, is agnostic to the non-uniformity. With assumption that all VCPUs have the same performance, it can carry out sub-optimal policies when allocating virtual resource for applications. Likewise, application runtime system can also make the same mistakes. Our focus in this paper is to understand the performance implications of the non-uniform VCPU-PCPU mapping in a virtualization system. Based on real measurements of a virtualization system with state of art multi-core processors running different commercial and emerging applications, we demonstrate that the presence of the non-uniform mapping has negative impacts on application’s performance predictability. This study aims to provide timely and practical insights on the problem of non-uniform VCPU mapping, when virtual machines being deployed and configured, in emerging cloud.     

On clustering in database servers for supporting mobile clients

The use of mobile computers is gaining popularity. There is an increasing trend in the number of users with laptops, PDAs, and smart phones. Access to information repositories in the future will be dominated by mobile clients rather than traditional “fixed” clients. These mobile clients download information by periodically connecting to repositories of data stored in either databases or file systems. Such mobile clients constitute a new and different kind of workload and exhibit a different access pattern than seen in traditional client server systems. Though file systems have been modified to handle clients that can download information, disconnect, and later reintegrate, databases have not been redesigned to accommodate mobile clients. There is a need to support mobile clients in the context of client server databases. This paper is about organizing the database server to take into consideration the access patterns of mobile clients. We propose the concept of hoard attributes which capture these access patterns. Three different techniques for organizing data on the server based on the hoard attribute are presented. We argue that each technique is suited for a particular workload. The workload is a combination of requests from mobile clients and traditional clients. This reorganization also allows us to address issues of concurrency control, disconnection and replica control in mobile databases. We present simulation results that show the performance of server reorganization using hoard attributes. We also provide an elaborate discussion of issues resulting from this reorganization in this new paradigm taking into account both mobile and traditional clients. This revised version was published online in August 2006 with corrections to the Cover Date.     

Middleware support for many-task computing

Many-task computing aims to bridge the gap between two computing paradigms, high throughput computing and high performance computing. Many-task computing denotes high-performance computations comprising multiple distinct activities, coupled via file system operations. The aggregate number of tasks, quantity of computing, and volumes of data may be extremely large. Traditional techniques found in production systems in the scientific community to support many-task computing do not scale to today’s largest systems, due to issues in local resource manager scalability and granularity, efficient utilization of the raw hardware, long wait queue times, and shared/parallel file system contention and scalability. To address these limitations, we adopted a “top-down” approach to building a middleware called Falkon, to support the most demanding many-task computing applications at the largest scales. Falkon (Fast and Light-weight tasK executiON framework) integrates (1) multi-level scheduling to enable dynamic resource provisioning and minimize wait queue times, (2) a streamlined task dispatcher able to achieve orders-of-magnitude higher task dispatch rates than conventional schedulers, and (3) data diffusion which performs data caching and uses a data-aware scheduler to co-locate computational and storage resources. Micro-benchmarks have shown Falkon to achieve over 15K+ tasks/s throughputs, scale to hundreds of thousands of processors and to millions of queued tasks, and execute billions of tasks per day. Data diffusion has also shown to improve applications scalability and performance, with its ability to achieve hundreds of Gb/s I/O rates on modest sized clusters, with Tb/s I/O rates on the horizon. Falkon has shown orders of magnitude improvements in performance and scalability than traditional approaches to resource management across many diverse workloads and applications at scales of billions of tasks on hundreds of thousands of processors across clusters, specialized systems, Grids, and supercomputers. Falkon’s performance and scalability have enabled a new class of applications called Many-Task Computing to operate at previously so-believed impossible scales with high efficiency.