Non-blocking Operations Research Articles

Optimal and scalable entanglement distribution over crossbar quantum networks.

Crossbar networks are a cornerstone of network architectures, capable of operating both as standalone interconnections or as integral switching components in complex, multi-stage systems. The main advantages of crossbar networks are their non-blocking operation and unparalleled minimal latency. With the advent of large scale quantum networks, crossbars might be an important asset towards the Quantum Internet. This study proposes a solution for the problem of distributing entanglement within crossbar quantum networks. Entangled particles are a consumable resource in quantum networks, and are being used by most quantum protocols. By ensuring that nodes within quantum networks are being supplied with entanglement, the reliability and efficiency of the network is maintained. By providing an efficient, scalable framework that can be used to achieve optimal entanglement distribution within crossbar quantum networks, this study offers a theoretical achievement which can be also used for enhancing quantum network performance. An algorithm for selecting an optimal entanglement distribution configuration is proposed and fully tested on realistic possible configurations.

Complementing JavaScript in High-Performance Node.js and Web Applications with Rust and WebAssembly

We examine whether the novel systems programming language named Rust can be utilized alongside JavaScript in Node.js and Web-based applications development. The paper describes how JavaScript can be used as a high-level scripting language in combination with Rust in place of C++ in order to realize efficiency and be free of race conditions as well as memory-related software issues. Furthermore, we conducted stress tests in order to evaluate the performance of the proposed architecture in various scenarios. Rust-based implementations were able to outperform JS by 1.15 by over 115 times across the range of measurements and overpower Node.js’s concurrency model by 14.5 times or more without the need for fine-tuning. In Web browsers, the single-thread WebAssembly implementation outperformed the respective pure JS implementation by about two to four times. WebAssembly executed inside of Chromium compared to the equivalent Node.js implementations was able to deliver 93.5% the performance of the single-threaded implementation and 67.86% the performance of the multi-threaded implementation, which translates to 1.87 to over 24 times greater performance than the equivalent manually optimized pure JS implementation. Our findings provide substantial evidence that Rust is capable of providing the low-level features needed for non-blocking operations and hardware access while maintaining high-level similarities to JavaScript, aiding productivity.

Callback-based completion notification using MPI Continuations

Asynchronous programming models (APM) are gaining more and more traction, allowing applications to expose the available concurrency to a runtime system tasked with coordinating the execution. While MPI has long provided support for multi-threaded communication and non-blocking operations, it falls short of adequately supporting APMs as correctly and efficiently handling MPI communication in different models is still a challenge. We have previously proposed an extension to the MPI standard providing operation completion notifications using callbacks, so-called MPI Continuations. This interface is flexible enough to accommodate a wide range of different APMs.In this paper, we present an extension to the previously described interface that allows for finer control of the behavior of the MPI Continuations interface. We then present some of our first experiences in using the interface in the context of different applications, including the NAS parallel benchmarks, the PaRSEC task-based runtime system, and a load-balancing scheme within an adaptive mesh refinement solver called ExaHyPE. We show that the interface, implemented inside Open MPI, enables low-latency, high-throughput completion notifications that outperform solutions implemented in the application space.

Wide-Sense Nonblocking Converting-Space-Converting Switching Node Architecture Under XsVarSWITCH Control Algorithm

In this study, we consider wide-sense nonblocking operation of a converting-space-converting switching fabric. We propose a routing strategy based on the functional decomposition of second-stage switches, called the XsVarSWITCH routing strategy. This routing strategy is suitable for all kinds of three-stage switching fabrics operating in a slotted mode, regardless of frequency, time, or other types. We consider in this study a three-stage switching fabric. The first and third stages comprise switches with converting capability, while the middle stage is composed of space switches. The proposed routing strategy can reduce the number of required center-stage switches by almost 94% in some cases, as compared to the number of middle-stage switches under strict-sense nonblocking conditions.

Design and Application of a Distribution Network Phasor Data Concentrator

The wide area measurement system (WAMS) based on synchronous phasor measurement technology has been widely used in power transmission grids to achieve dynamic monitoring and control of the power grid. At present, to better realize real-time situational awareness and control of the distribution network, synchronous phasor measurement technology has been gradually applied to the distribution network, such as the application of micro multifunctional phasor measurement units (μMPMUs). The distribution network phasor data concentrator (DPDC), as a connection node between the μMPMUs and the main station, is also gaining more attraction. This paper first analyzes the communication network structure of DPDCs and μMPMUs and compares and analyzes the differences in the installation locations, functions, communication access methods and communication protocols of the phasor technology devices of the distribution network and the transmission network. It is pointed out that DPDCs not only need the functions of data collection, storage, and forwarding like transmission network PDCs, but also should be able to access more μMPMUs, and can aggregate the phasor data of the same time scale from μMPMUs by different communication methods. The communication protocol selected by DPDC should be expanded to support remote control, telemetry, fault diagnosis and other functions of distribution automation. The application requirements of DPDCs are clarified, and the key indicators of DPDCs are given as a method to evaluate the basic performance of DPDCs. Then, to address the problems of more μMPMU access, abnormal communication, and data collection with different delays that DPDC encountered, a DPDC that considers multiple communication methods is designed. Based on the Linux system and the libuv library, the DPDC is designed with event-driven mechanism and structured programming, runs multiple threads to implement multitasking, and invokes callbacks to perform asynchronous non-blocking operations. The DPDC test system and test methods are designed. The performance of the designed DPDC is evaluated through the test and the test results are analyzed. Lastly, its real-world application is disclosed, which further confirmed the value of our DPDC.

RootJS: Node.js Bindings for ROOT 6

We present rootJS, an interface making it possible to seamlessly integrate ROOT 6 into applications written for Node.js, the JavaScript runtime platform increasingly commonly used to create high-performance Web applications. ROOT features can be called both directly from Node.js code and by JIT-compiling C++ macros. All rootJS methods are invoked asynchronously and support callback functions, allowing non-blocking operation of Node.js applications using them. Last but not least, our bindings have been designed to platform-independent and should therefore work on all systems supporting both ROOT 6 and Node.js.Thanks to rootJS it is now possible to create ROOT-aware Web applications taking full advantage of the high performance and extensive capabilities of Node.js. Examples include platforms for the quality assurance of acquired, reconstructed or simulated data, book-keeping and e-log systems, and even Web browser-based data visualisation and analysis.

Static deadlock detection for asynchronous C# programs

Asynchronous programming is a standard approach for designing responsive applications. Modern languages such as C# provide async/await primitives for the disciplined use of asynchrony. In spite of this, programs can deadlock because of incorrect use of blocking operations along with non-blocking (asynchronous) operations. While developers are aware of this problem, there is no automated technique to detect deadlocks in asynchronous programs. We present a novel representation of control flow and scheduling of asynchronous programs, called continuation scheduling graph and formulate necessary conditions for a deadlock to occur in a program. We design static analyses to construct continuation scheduling graphs of asynchronous C# programs and to identify deadlocks in them. We have implemented the static analyses in a tool called DeadWait. Using DeadWait, we found 43 previously unknown deadlocks in 11 asynchronous C# libraries. We reported the deadlocks to the library developers. They have confirmed and fixed 40 of them.

Wide-sense nonblocking elastic optical switch

In this paper, we address the problem of wide-sense nonblocking operation of the elastic optical switch. The switch consists of waveband-converting switches in input and output stages, while in the center stage there is only one wavelength-selective space switch without wavelength-conversion capabilities. Two control algorithms are proposed, namely, 2-split and 3-split. For these algorithms we derived wide-sense nonblocking conditions and compared them with strict-sense nonblocking ones. The results show that the 3-split algorithm reduces the required number of FSUs to up to almost 70% of the FSUs required in the strict-sense nonblocking switching fabrics. Application of the proposed elastic optical switch in optical datacenter networks is also proposed.

Petri-net-based robust supervisory control of automated manufacturing systems

Supervisory control that ensures deadlock-free and nonblocking operation has been an active research area of manufacturing engineering. So far, most of deadlock control policies in the existing literature assume that allocated resources are reliable. Additionally, a large number of methods are for systems of simple sequential processes with resources (S3PRs), where a part uses only one copy of one resource at each processing step. In contrast, we investigate the automated manufacturing systems (AMSs) that can be modeled by a class of Petri nets, namely S*PUR. S*PUR is a generalization of the S*PR Petri net model, while S*PR is a superclass of S3PR. This work addresses the robust supervision for deadlock avoidance in S*PUR. Specifically, we take into account unreliable resources that may break down while working or being in idle, and the considered AMSs allow the use of multiple copies of different resources per operation stage. Our objective is to control the system so that: 1) when there are breakdowns, the system can continue producing parts of some types whose production does not need any failed resources; and 2) given the correction of all faults, it is possible to complete all the on-going part instances remaining in the system. We illustrate the characteristics of a desired supervisor through several examples, define the corresponding properties of robustness, and develop a control policy that satisfies such properties.

Strict-sense nonblocking networks with k degrees of freedom

In this paper we extended the well known conditions for strict-sense nonblocking operation of the three-stage Clos switching network to the case in which k multiplexing levels (called later in this paper as k degrees of freedom) are used on input, output, and interstage links. Up till now only the cases with one multiplexing level were considered in the literature—time-division or wavelength-division. Current research shows that transmission systems with many multiplexing levels will be used in the future to increase available transmission speed in telecommunication networks. There will be also a need for switching connecting paths going through such multiplexed links in large capacity optical cross-connect systems, routers, or data center networks. In this paper we proved the strict-sense nonblocking conditions for the three-stage Clos network with k degrees of freedom. We also analyzed the cost of switching fabrics of selected capacities and with three degrees of freedom in terms of the required number of elements.

SyncAS: A Virtual Block Approach to Tame Asynchronous Programming

Asynchronous programming has been widely adopted in domains such as Web development. This programming style usually uses callback methods, non-blocking operations, allowing high responsive user interactions even if an application works without multi-threading. However, this style requires the uncoupling of a module into two sub-modules at least, which are not intuitively connected by a callback method. The separation of modules spurs the birth of other issues: callback spaghetti and callback hell. This paper proposes a virtual block approach to address the previous two issues. This approach enables a programmer to virtually block a program execution and restart it at arbitrary points in the program. As a result, programmers do not need to uncouple a module even if non-blocking operations are adopted; therefore, callback dependencies disappear. Using aspect-oriented programming, this approach uses aspects to control the execution of a program in an oblivious manner. As a consequence, programmers do not need to be concerned whether pieces of code use blocking or non-blocking operations. We implement a proof-of-concept for this approach, called SyncAS, for ActionScript3. In addition, we apply our proposal to a toy application and conduct experiments to show its modular application, flexibility, and performance.

T-NBC: Transparent Non-blocking MPI Collective Operations

Without modifying MPI applications,transparently translating MPI collective operations into non-blocking ones in communication libraries can overlap collective communication with the computation following the operations and benefit most current applications.In applications,the following computation includes communication-unrelated computation(CURC) and communication-related computation(CRC).CURC is easier to overlap with collective communication;however,CRC need access communication data and is more difficult to overlap with collective communication.In the paper,we propose transparent non-blocking collective operations(T-NBC).It can obtain the overlap between collective communication and following communication.Besides the overlap with CURC,it improves the overlap with CRC by transmitting collective messages with different priorities according to their accessed sequence in applications.Evaluations of micro-benchmark demonstrate that a large potential overlap between collective communication and following computation can be obtained.In FT(Fourier Transform) and IS(Integer Sort) of NPB(NAS Parallel Benchmarks),even following computation dominated by CRC,a large portion of collective communication is overlapped.Their performance is respectively improved by 5% and 36%.

CMOS compatible reconfigurable filter for high bandwidth non-blocking operation

We design, fabricate and characterize a CMOS-compatible, Mach-Zehnder-coupled, second-order-microring-resonator filter with large Free Spectral Range and demonstrate non-blocking thermo-optical filter reconfiguration. The device consists of 10-μm radius silicon microring resonators, with an FSR equivalent to that of a structure consisting of 5-μm radii microrings. The structure is reconfigurable over an 8.5 nm range without blocking other channels in the network.

Predictable performance and high query concurrency for data analytics

Conventional data warehouses employ the query-at-a-time model, which maps each query to a distinct physical plan. When several queries execute concurrently, this model introduces contention and thrashing, because the physical plans--unaware of each other--compete for access to the underlying I/O and computation resources. As a result, while modern systems can efficiently optimize and evaluate a single complex data analysis query, their performance suffers significantly and can be highly erratic when multiple complex queries run at the same time. We present in this paper Cjoin, a new design that substantially improves throughput in large-scale data analytics systems processing many concurrent join queries. In contrast to the conventional query-at-a-time model our approach employs a single physical plan that shares I/O, computation, and tuple storage across all in-flight join queries. We use an always pipeline of non-blocking operators, managed by a controller that continuously examines the current query mix and optimizes the pipeline on the fly. Our design enables data analytics engines to scale gracefully to large data sets, provide predictable execution times, and reduce contention. We implemented Cjoin as an extension to the PostgreSQL DBMS. This prototype outperforms conventional commercial systems by an order of magnitude for tens to hundreds of concurrent queries.

MPSoC를 위한 저비용 하드웨어 MPI 유닛 설계

본 논문에선 분산 메모리 아키텍처를 사용하는 멀티프로세서 시스템에서 메시지 전달을 지원하는 하드웨어 MPI(Message Passing Interface) 유닛을 설계하였다 데이터 전송 동기화 및 데이터 전송, 완료까지의 과정을 하드웨어 MPI 유닛이 담당하여 동기화에 따른 오버헤드를 경감시켰다. 또한 동기화 메시지를 저장 관리하는 요청 큐(Request Queue), 준비 큐(Ready Queue), 예약 큐(Reserve Queue)를 내장하여 병렬적으로 입력받은 동기화 메시지를 관리하고 비순차적 종료(out of order completion)을 지원한다. BMF(Bus Functional Medel)을 제작해 제안한 구조에서의 전송 대역폭 성능을 확인한 결과 다대다 통신에서 25% 이상의 성능 향상이었음을 확인할 수 있었다. 그 후 HDL로 기술된 하드웨어를 Magnachip 0.18 공정 라이브러리에서 합성하였으며 프로토 타입 chip으로 제작하였다. 제안한 MPI 유닛은 전체 칩 사이즈의 1% 이하의 크기로 높은 성능 향상을 기대할 수 있어, 저비용 설계와 확장성 측면에서 임베디드 MPSoC(Multi-Processor System-on-Chip)의 전체적인 성능을 높이는데 유용하다. In this paper, we propose a novel hardware MPI(Message Passing Interface) unit which supports message passing in multiprocessor system which use distributed memory architecture. MPI Hardware unit processes data synchronization, transmission and completion, and it supports processor non-blocking operation so it reduces overhead according to synchronization. Additionally, MPI hardware unit combines ready entry, request entry, reserve entry which save and manage the synchronized messages and performs the multiple outstanding issue and out of order completion. According to BFM(Bus Functional Model) simulation result, the performance is increased by 25% on many to many communication. After we designed MPI unit using HDL, with synopsys design compiler we synthesized, and for synthesis library we used MagnaChip <TEX>$0.18{\mu}m$</TEX>. And then we making prototype chip. The proposed message transmission interface hardware shows high performance for its increase in size. Thus, as we consider low-cost design and scalability, MPI hardware unit is useful in increasing overall performance of embedded MPSoC(Multi-Processor System-on-Chip).

Wide-sense nonblocking multicast logd (N, m, p) switching networks under the window algorithm

The generalized multiplane wide-sense nonblocking switching network for multicast connections is presented in this paper. This switching network is based on the well-known baseline network. Up to now, wide-sense nonblocking operations for multicast connections have been considered for the following switching networks which are based on the baseline network: multiplane switching networks composed of 2 × 2 switching elements (called log2 (N, 0, p) networks), multiplane switching networks composed of 2 x 2 switching elements and with m extra stages (called log2 (N, m, p) networks), and multiplane generalized switching networks composed of d x d switching elements (called logd (N, 0, p) networks). The control algorithm under which these networks are wide-sense nonblocking for multicast connections is called the window algorithm. In this paper, multiplane switching networks composed of dxd switching elements with m extra stages (called logd (N, m, p) networks) are considered and in this way the last link in the studies of wide-sense nonblocking multicast multiplane switching networks under blocking window algorithms is presented. In this article, the path-searching algorithm is adopted for the general case d ? 2 and the wide-sense nonblocking conditions for switching networks for different sizes of the blocking window and for different numbers of extra stages are derived. The cost of different multiplane switching networks is compared in terms of the number of planes and the number of crosspoints.

Two Parallel Algorithms for a Mass Transfer Simulation of Magnetic Nanoparticles

We present a comparative study on the performance of two parallel algorithms for simulating mass transfer of weakly magnetic nanoparticles during the process of High Gradient Magnetic Separation (HGMS). The dynamics of mass transfer is investigated statistically in term of particle volume concentration and is described by the continuity equation which is solved numerically using the finite-difference. For parallelization, the concentration data are divided into equal parts that are distributed to a group of parallel processes. Parallel computations are performed by using two communication schemes of MPI, the pair-wise blocking and non-blocking operations. We compare the performance of both schemes interms of parallel speedup, efficiency, and communication overhead. The results show that parallel simulation using the non-blocking communication has better performance than the blocking communication and also shows the better scalability with increasing number of processes.

A scalable, predictable join operator for highly concurrent data warehouses

Conventional data warehouses employ the query-at-a-time model, which maps each query to a distinct physical plan. When several queries execute concurrently, this model introduces contention, because the physical plans---unaware of each other---compete for access to the underlying I/O and computation resources. As a result, while modern systems can efficiently optimize and evaluate a single complex data analysis query, their performance suffers significantly when multiple complex queries run at the same time. We describe an augmentation of traditional query engines that improves join throughput in large-scale concurrent data warehouses. In contrast to the conventional query-at-a-time model, our approach employs a single physical plan that can share I/O, computation, and tuple storage across all in-flight join queries. We use an "always-on" pipeline of non-blocking operators, coupled with a controller that continuously examines the current query mix and performs run-time optimizations. Our design allows the query engine to scale gracefully to large data sets, provide predictable execution times, and reduce contention. In our empirical evaluation, we found that our prototype outperforms conventional commercial systems by an order of magnitude for tens to hundreds of concurrent queries.

Nonblocking Hierarchical Control of Decentralized Discrete Event Systems

This contribution investigates the hierarchical control of decentralized discrete event systems (DES) that are synchronized by shared events. A hierarchical control architecture providing hierarchical consistency is introduced. Moreover, it allows for composition of decentralized subsystems on the high-level of the hierarchy and hence reduces the computational complexity of supervisory control synthesis for language inclusion specifications. In this context, a crucial issue is the nonblocking operation of the overall system. Our main theorem identifies sufficient conditions for this desirable property.

THINC

Rapid improvements in network bandwidth, cost, and ubiquity combined with the security hazards and high total cost of ownership of personal computers have created a growing market for thin-client computing. We introduce THINC, a virtual display architecture for high-performance thin-client computing in both LAN and WAN environments. THINC virtualizes the display at the device driver interface to transparently intercept application display commands and translate them into a few simple low-level commands that can be easily supported by widely used client hardware. THINC's translation mechanism efficiently leverages display semantic information through novel optimizations such as offscreen drawing awareness, native video support, and server-side screen scaling. This is integrated with an update delivery architecture that uses shortest command first scheduling and non-blocking operation. THINC leverages existing display system functionality and works seamlessly with unmodified applications, window systems, and operating systems.We have implemented THINC in an X/Linux environment and compared its performance against widely used commercial approaches, including Citrix MetaFrame, Microsoft RDP, GoToMyPC, X, NX, VNC, and Sun Ray. Our experimental results on web and audio/video applications demonstrate that THINC can provide up to 4.8 times faster web browsing performance and two orders of magnitude better audio/video performance. THINC is the only thin client capable of transparently playing full-screen video and audio at full frame rate in both LAN and WAN environments. Our results also show for the first time that thin clients can even provide good performance using remote clients located in other countries around the world.