Sharif Digital Repository

Recently, Grid infrastructures have provided wide integrated use of resources. DotGrid intends to introduce required Grid services and toolkits that are implemented as a layer wrapped over the existing operating systems. Our DotGrid has been developed based on Microsoft .NET in Windows and MONO .NET in Linux and UNIX. Using DotGrid APIs, Grid middlewares and applications can be implemented easily. We evaluated our DotGrid capabilities by implementing some applications including a grid-based distributed cryptographic engine and also a typical computational problem. © 2006 IEEE  

This paper presents the SkipTree, a new balanced, distributed data structure for storing data with multidimensional keys in a peer-to-peer network. The SkipTree supports range queries as well as single point queries which are routed in O(log n) hops. SkipTree is fully decentralized with each node being connected to O(logn) other nodes. The memory usage for maintaining the links at each node is O(log n log log n) on average and O(log2 n) in the worst case. Load balance is also guaranteed to be within a constant factor. © Springer-Verlag Berlin Heidelberg 2005  

How can we optimally trade extra computing power to reduce the communication load in distributed computing? We answer this question by characterizing a fundamental tradeoff between computation and communication in distributed computing, i.e., the two are inversely proportional to each other. More specifically, a general distributed computing framework, motivated by commonly used structures like MapReduce, is considered, where the overall computation is decomposed into computing a set of “Map” and “Reduce” functions distributedly across multiple computing nodes. A coded scheme, named “Coded Distributed Computing” (CDC), is proposed to demonstrate that increasing the computation load of the... 

In cloud computing, services play key roles. Services are well defined and autonomous components. Nowadays, the demand of using Fuzzy inference as a service is increasing in the domain of complex and critical systems. In such systems, along with the development of the software, the cost of detecting and fixing software defects increases. Therefore, using formal methods, which provide clear, concise, and mathematical interpretation of the system, is crucial for the design of these Fuzzy systems. To obtain this goal, we introduce the Fuzzy Inference Cloud Service (FICS) and propose a novel discipline for formal modeling of the FICS. The FICS provides the service of Fuzzy inference to the... 

It is of fundamental importance to find algorithms obtaining optimal performance for learning of statistical models in distributed and communication limited systems. Aiming at characterizing the optimal strategies, we consider learning of Gaussian Processes (GP) in distributed systems as a pivotal example. We first address a very basic problem: how many bits are required to estimate the inner-products of some Gaussian vectors across distributed machines? Using information theoretic bounds, we obtain an optimal solution for the problem which is based on vector quantization. Two suboptimal and more practical schemes are also presented as substitutes for the vector quantization scheme. In... 

This paper proposes a distributed attack detection and mitigation technique based on distributed estimation over a multi-agent network, where the agents take partial system measurements susceptible to (possible) biasing attacks. In particular, we assume that the system is not locally observable via the measurements in the direct neighborhood of any agent. First, for performance analysis in the attack-free case, we show that the proposed distributed estimation is unbiased with bounded mean-square deviation in steady-state. Then, we propose a residual-based strategy to locally detect possible attacks at agents. In contrast to the deterministic thresholds in the literature assuming an upper... 

Federated learning (FL) and split learning (SL) are state-of-the-art distributed machine learning techniques to enable machine learning without accessing raw data on clients or end devices. However, their comparative training performance under real-world resource-restricted Internet of Things (IoT) device settings, e.g., Raspberry Pi, remains barely studied, which, to our knowledge, have not yet been evaluated and compared, rendering inconvenient reference for practitioner. This work firstly provides empirical comparisons of FL and SL in real-world IoT settings regarding learning performance and on-device execution overhead. Our analyses demonstrate that the learning performance of SL is... 

There is growing interest in applying distributed machine learning to edge computing, forming federated edge learning. Federated edge learning faces non-i.i.d. and heterogeneous data, and the communication between edge workers, possibly through distant locations and with unstable wireless networks, is more costly than their local computational overhead. In this work, we propose ${{sf DONE}}$DONE, a distributed approximate Newton-type algorithm with fast convergence rate for communication-efficient federated edge learning. First, with strongly convex and smooth loss functions, ${{sf DONE}}$DONE approximates the Newton direction in a distributed manner using the classical Richardson iteration... 

There are a vast number of cloud service providers, which offer virtual machines (VMs) with different configurations. From the companies perspective, an appropriate selection of VMs is an important issue, as the proper service selection leads to improved productivity, higher efficiency, and lower cost. An effective service selection cannot be done without a systematic approach due to the modularity of requests, the conflicts between requirements, and the impact of network parameters. In this paper, we introduce an innovative framework, called PCA, to solve service selection problem in the hybrid environment of peer-assisted, public, and private clouds. PCA detects the conflicts between the... 

How can we optimally trade extra computing power to reduce the communication load in distributed computing? We answer this question by characterizing a fundamental tradeoff between computation and communication in distributed computing, i.e., the two are inversely proportional to each other. More specifically, a general distributed computing framework, motivated by commonly used structures like MapReduce, is considered, where the overall computation is decomposed into computing a set of “Map” and “Reduce” functions distributedly across multiple computing nodes. A coded scheme, named “coded distributed computing” (CDC), is proposed to demonstrate that increasing the computation load of the... 

In this paper, learning of tree-structured Gaussian graphical models from distributed data is addressed. In our model, samples are stored in a set of distributed machines where each machine has access to only a subset of features. A central machine is then responsible for learning the structure based on received messages from the other nodes. We present a set of communication-efficient strategies, which are theoretically proved to convey sufficient information for reliable learning of the structure. In particular, our analyses show that even if each machine sends only the signs of its local data samples to the central node, the tree structure can still be recovered with high accuracy. Our... 

In this paper, we argue that in many basic algorithms for machine learning, including support vector machine (SVM) for classification, principal component analysis (PCA) for dimensionality reduction, and regression for dependency estimation, we need the inner products of the data samples, rather than the data samples themselves.Motivated by the above observation, we introduce the problem of private inner product retrieval for distributed machine learning, where we have a system including a database of some files, duplicated across some non-colluding servers. A user intends to retrieve a subset of specific size of the set of the inner product of every pair of data items in the database with... 

In this paper, learning of tree-structured Gaussian graphical models from distributed data is addressed. In our model, samples are stored in a set of distributed machines where each machine has access to only a subset of features. A central machine is then responsible for learning the structure based on received messages from the other nodes. We present a set of communication-efficient strategies, which are theoretically proved to convey sufficient information for reliable learning of the structure. In particular, our analyses show that even if each machine sends only the signs of its local data samples to the central node, the tree structure can still be recovered with high accuracy. Our... 

In this paper, learning of tree-structured Gaussian graphical models from distributed data is addressed. In our model, samples are stored in a set of distributed machines where each machine has access to only a subset of features. A central machine is then responsible for learning the structure based on received messages from the other nodes. We present a set of communication-efficient strategies, which are theoretically proved to convey sufficient information for reliable learning of the structure. In particular, our analyses show that even if each machine sends only the signs of its local data samples to the central node, the tree structure can still be recovered with high accuracy. Our...