Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userCrux: Locality-Preserving Distributed Services
165
0
0.0
(
0
)
Added by
Bryan Ford
Publication date
2014
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Distributed systems achieve scalability by distributing load across many machines, but wide-area deployments can introduce worst-case response latencies proportional to the networks diameter. Crux is a general framework to build locality-preserving distributed systems, by transforming an existing scalable distributed algorithm A into a new locality-preserving algorithm ALP, which guarantees for any two clients u and v interacting via ALP that their interactions exhibit worst-case response latencies proportional to the network latency between u and v. Crux builds on compact-routing theory, but generalizes these techniques beyond routing applications. Crux provides weak and strong consistency flavors, and shows latency improvements for localized interactions in both cases, specifically up to several orders of magnitude for weakly-consistent Crux (from roughly 900ms to 1ms). We deployed on PlanetLab locality-preservi
rate research
Read More
State-of-the-art distributed in-memory datastores (FaRM, FaSST, DrTM) provide strongly-consistent distributed transactions with high performance and availability. Transactions in those systems are fully general; they can atomically manipulate any set of objects in the store, regardless of their location. To achieve this, these systems use complex distributed transactional protocols. Meanwhile, many workloads have a high degree of locality. For such workloads, distributed transactions are an overkill as most operations only access objects located on the same server -- if sharded appropriately. In this paper, we show that for these workloads, a single-node transactional protocol combined with dynamic object re-sharding and asynchronously pipelined replication can provide the same level of generality with better performance, simpler protocols, and lower developer effort. We present Zeus, an in-memory distributed datastore that provides general transactions by acquiring all objects involved in the transaction to the same server and executing a single-node transaction on them. Zeus is fault-tolerant and strongly-consistent. At the heart of Zeus is a reliable dynamic object sharding protocol that can move 250K objects per second per server, allowing Zeus to process millions of transactions per second and outperform more traditional distributed transactions on a wide range of workloads that exhibit locality.
Distributed Services Architecture with support for mobile agents between services, offer significantly improved communication and computational flexibility. The uses of agents allow execution of complex operations that involve large amounts of data to be processed effectively using distributed resources. The prototype system Distributed Agents for Mobile and Dynamic Services (DIAMOnDS), allows a service to send agents on its behalf, to other services, to perform data manipulation and processing. Agents have been implemented as mobile services that are discovered using the Jini Lookup mechanism and used by other services for task management and communication. Agents provide proxies for interaction with other services as well as specific GUI to monitor and control the agent activity. Thus agents acting on behalf of one service cooperate with other services to carry out a job, providing inter-operation of loosely coupled services in a semi-autonomous way. Remote file system access functionality has been incorporated by the agent framework and allows services to dynamically share and browse the file system resources of hosts, running the services. Generic database access functionality has been implemented in the mobile agent framework that allows performing complex data mining and processing operations efficiently in distributed system. A basic data searching agent is also implemented that performs a query based search in a file system. The testing of the framework was carried out on WAN by moving Connectivity Test agents between AgentStations in CERN, Switzerland and NUST, Pakistan.
Cloud Computing (CC) is a model for enabling on-demand access to a shared pool of configurable computing resources. Testing and evaluating the performance of the cloud environment for allocating, provisioning, scheduling, and data allocation policy have great attention to be achieved. Therefore, using cloud simulator would save time and money, and provide a flexible environment to evaluate new research work. Unfortunately, the current simulators (e.g., CloudSim, NetworkCloudSim, GreenCloud, etc..) deal with the data as for size only without any consideration about the data allocation policy and locality. On the other hand, the NetworkCloudSim simulator is considered one of the most common used simulators because it includes different modules which support needed functions to a simulated cloud environment, and it could be extended to include new extra modules. According to work in this paper, the NetworkCloudSim simulator has been extended and modified to support data locality. The modified simulator is called LocalitySim. The accuracy of the proposed LocalitySim simulator has been proved by building a mathematical model. Also, the proposed simulator has been used to test the performance of the three-tire data center as a case study with considering the data locality feature.
The GLEON Research And PRAGMA Lake Expedition -- GRAPLE -- is a collaborative effort between computer science and lake ecology researchers. It aims to improve our understanding and predictive capacity of the threats to the water quality of our freshwater resources, including climate change. This paper presents GRAPLEr, a distributed computing system used to address the modeling needs of GRAPLE researchers. GRAPLEr integrates and applies overlay virtual network, high-throughput computing, and Web service technologies in a novel way. First, its user-level IP-over-P2P (IPOP) overlay network allows compute and storage resources distributed across independently-administered institutions (including private and public clouds) to be aggregated into a common virtual network, despite the presence of firewalls and network address translators. Second, resources aggregated by the IPOP virtual network run unmodified high-throughput computing middleware (HTCondor) to enable large numbers of model simulations to be executed concurrently across the distributed computing resources. Third, a Web service interface allows end users to submit job requests to the system using client libraries that integrate with the R statistical computing environment. The paper presents the GRAPLEr architecture, describes its implementation and reports on its performance for batches of General Lake Model (GLM) simulations across three cloud infrastructures (University of Florida, CloudLab, and Microsoft Azure).
This paper studies the problem of code symbol availability: a code symbol is said to have $(r, t)$-availability if it can be reconstructed from $t$ disjoint groups of other symbols, each of size at most $r$. For example, $3$-replication supports $(1, 2)$-availability as each symbol can be read from its $t= 2$ other (disjoint) replicas, i.e., $r=1$. However, the rate of replication must vanish like $frac{1}{t+1}$ as the availability increases. This paper shows that it is possible to construct codes that can support a scaling number of parallel reads while keeping the rate to be an arbitrarily high constant. It further shows that this is possible with the minimum distance arbitrarily close to the Singleton bound. This paper also presents a bound demonstrating a trade-off between minimum distance, availability and locality. Our codes match the aforementioned bound and their construction relies on combinatorial objects called resolvable designs. From a practical standpoint, our codes seem useful for distributed storage applications involving hot data, i.e., the information which is frequently accessed by multiple processes in parallel.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
