Computational models are an essential tool for the design, characterization, and discovery of novel materials. Computationally hard tasks in materials science stretch the limits of existing high-performance supercomputing centers, consuming much of their resources for simulation, analysis, and data processing. Quantum computing, on the other hand, is an emerging technology with the potential to accelerate many of the computational tasks needed for materials science. In order to do that, the quantum technology must interact with conventional highperformance computing in several ways: approximate results validation, identification of hard problems, and synergies in quantum-centric supercomputing. In this paper, we provide a perspective on how quantum-centric supercomputing can help address critical computational problems in materials science, the challenges to face in order to solve representative use cases, and new suggested directions.

Quantum computing has the potential to revolutionize the way we solve complex problems in various fields. Its unique features, such as superposition and entanglement, allow for speedup in certain computations compared to classical computers. In this paper, we show that incorporating quantum computing into a heterogeneous computation architecture is possible and potentially can greatly enhance the overall performance and efficiency of solving a wide range of problems. We propose a heterogeneous computation architecture that integrates quantum computing and classical computing. We describe how this architecture can be implemented using widely available opensource software and provide a reference implementation. We also discuss potential benefits and challenges. Our goal is to highlight the importance of seamless inclusion of quantum computing in the future of computation and how it can be achieved with tools we already have.

The security and privacy of cyber-physical systems (CPS) has been the subject of a lot of concern in the recent past especially in the context of the safety of critical infrastructure such as the power grid, transportation, and manufacturing environment. We envision that ensuring the security and privacy of these systems is really part of designing a resilient cyber-physical system. The grand challenge is really to provide continuous assurance of the operating objectives. In order to achieve continuous assurance, we believe fundamental research efforts need to be devoted to the challenges faced by fine-grained isolation, real-world aware, risk-adjusted resource allocation, and closing the loop.

The increasing demand for broadband services in the Internet will create a network bandwidth shortage and severely limit the capability to broadcast rich media in the Internet. A key impediment for broadband deployment is that there is no universal support for IP multicast in the Internet, because of several difficulties ranging from scalability problems to the accounting of network traffic and the enforcement of quality of service to subscribers. Also, measures that regulate the amount of traffic that can be generated from a given publisher such as committed access rate (CAR) agreements will severely limit the amount of data that can be published by a given node. In this paper, a new framework is discussed which dynamically builds an application level multicast network employing the peer nodes of a given application as network routers, thus greatly reducing the network bandwidth requirements for data broadcast. The paper discusses the current measures of network traffic accounting currently employed by Internet service providers and shows how the peer to peer model can circumvent several of the limitations of the present day Internet.

Distributed interactive applications such as multiplayer games will become increasingly popular in wide area distributed systems. To provide the response time desired by users despite high and unpredictable communication latency in such systems, shared objects will be replicated orcached by clients that participate in the applications. Any updates to the shared objects will have to be disseminated to clients that actually use the objects to maintain consistency. We address the problem of efficient and scalable update dissemination in an environment where client interests can change dynamically and the number of multicast channels available for update dissemination is limited. We present a heuristic based algorithm that can group objectsand clients in a way that it handles limited bandwidth resources. We show that our algorithm can produce better results than several algorithms that have been developed in the past for update dissemination.

Predicted availability of broadband access will enable deployment of network Virtual Environments over the Internet. Online virtual worlds will require efficient run-time data replication solutions, This paper presents a Quality of Service (QoS) architecture for just-in-time data replication in network Virtual Environments. Quality of service is achieved by predicting the load and adapting to network traffic variations. Data is prefetched at the client based on network traffic estimates and viewpoint navigation prediction. QoS negotiation allows the server to control the network resources allocated per client. Experimental results show that QoS data replication can be implemented with reasonably small network and server overload.

This paper discusses orthogonality in layered multicast schemes, and presents a general framework for the implementation of layered multicast. This framework can be applied to standard IP multicast or to any other multicast technology capable of transmitting a data stream to a group of subscriber nodes. The multirate transmission is accomplished by a multicast group management facility which maps the frames of the multimedia stream into a number of multicast groups according to different data resolutions. The subscriber nodes can then directly control the data rate of the stream that they receive by joining different multicast groups, thus eliminating the need for constant data rate negotiation with the sender node.

In this paper, we present the design and implementation of NetBazaar, which is a distributed, federated electronic trading system (Marketplace) for buying and selling network resources and services and information products and services distributed across the Internet. The trading system provides mechanisms for suppliers to advertise information about their services and attribute-value pairs, and for consumers to query for information about service offerings by the suppliers. In addition, the trading system offers services to perform the trades on behalf of the consumers or it offers the consumers with a list of suppliers to contact. In order to recover costs and profit, the trading system charges a small fee to the suppliers and consumers for every trade that occurs. The charges could vary depending on the complexity of the trade, such as the overheads of payment, transaction and contract enforcement. NetBazaar has been designed to support a variety of business models, pricing and market mechanisms, searching and matching algorithms, fast negotiation mechanisms for a high volume of trades, and distributed access for consumers and suppliers to the trading system. An initial version of NetBazaar has been implemented using CORBA and Java components.

Container-managed messaging (CMM) allows a Java component to communicate via messaging without having to manage the messaging infrastructure, similarly to how container-managed persistence (CMP) allows an EJB (Enterprise JavaBean) to have persistent data without managing access to a data store. In addition, messaging parameters (such as destinations, mode of interaction, time-outs, etc.) can be defined declaratively in a deployment descriptor. The programming model used in CMM resembles that of CORBA messaging with the exception that it is in an anonymous and declarative fashion

Objects were introduced as programming constructs that encapsulate data and methods. The goal was to foster software reuse and simplify the developer's concept of how a task was implemented. The developer need only know the interfaces to an object to use ifs functionality. Distributed objects simplified conceptualization further by removing the need to know the locality of an object Clients invoked methods on distributed objects as if the objects existed in the client's process. Beyond this location transparency, the need arose for distributed objects to survive beyond the life of one client, to be able to support thousands or millions of clients, and to participate in transactions. To support scalability, persistence, and transactional semantics with no dependencies on platform or data store, "component models" were developed. In this paper we look at various component models, focusing on two: IBM's Component Broker and Sun's Enterprise JavaBeans(TM). We show that they augment each other and propose how Enterprise JavaBeans can use the additional functions of Component Broker to provide a scalable, transactional, and persistent environment to clients of both worlds.

The Frauenkirche was destroyed when Dresden was bombed by the Allied forces February 13-14, 1945. The church is now being reconstructed in an effort led by the Foundation for the Reconstruction of the Frauenkirche. The VRDECK software package developed at the IBM T. J. Watson Research Center was used to view and walk through a model created from the original church plans. A Polhemus tracker and a custom-built joystick using the Logitech 3D mouse were used for six-degree-of-freedom input to the application. The interactive fly-through of the church is in an immersive environment. One can navigate around the model wearing a head-mounted display, sitting in front of a standard monitor, looking at a stereo image produced on a stereo monitor, or standing before a projection screen displaying a stereo image of the scene. The system was developed for and exhibited at the IBM booth in the CeBIT fair in Hannover, Germany in March 1994 with funding from IBM Germany.

The design and operation of the Virtual Reality Distributed Environment and Construction Kit (VR-DECK) toolkit developed at IBM Research is reviewed. It provides a designer with a development environment while supporting distributed computing, multi-user capability, and a variety of I/O devices. Virtual worlds are built as collections of modules which communicate via events. Extensive run-time support in the form of extensive C++ class libraries insulates the application designer from the low-level system details such as networking, inter-module data transport, event queuing and matching, and I/O device communication. A library of pre-defined modules is provided for commonly used functions and devices. An X Window System graphical user interface is provided for aggregating modules into applications. The system enables a developer to focus on the design of the application rather than on systems and integration issues.<>