This paper presents a novel approach to high-performance video processing that avoids the use of ... more This paper presents a novel approach to high-performance video processing that avoids the use of special purpose hardware and achieves parallelism via a distributed system of high- performance workstations. An efficient network protocol allows mobile computers or mobile robots to remotely access the distributed video processing system over low-bandwidth wireless links. Despite the lack of special purpose hardware and expensive computing equipment, the system provides exceptional performance processing data at speeds that approach real-time. The novel contribution of our approach is the ability to process video in compressed format and efficiently distribute video data to a distributed system of workstations for processing. We describe the algorithms employed to achieve these features and demonstrate our approach by showing how it can be applied to two problem domains: multiple target tracking and video archive searching. We present experimental results that show that video data can be efficiently disseminated to a network or workstations with only a small (7%) increase in transfer time. We also show that processing can be performed efficiently (20 fps) directly on the compressed video without social purpose hardware.
... 2. [clouds] Susan Angebranndt, Richard L. hyde, Daphne Huetu Luong, and Nagendra Siravara. In... more ... 2. [clouds] Susan Angebranndt, Richard L. hyde, Daphne Huetu Luong, and Nagendra Siravara. Integrating audio and telephony in a distributed workstion environment. In USENIX '91 Conference Proceedings, pages 419-435. USENIX, June 1991. ... 21. [mcp-dcs] Raj Yavatkar. ...
This paper defines a new group communication model called concast communication. Being the counte... more This paper defines a new group communication model called concast communication. Being the counterpart to multicast, concast involves multiple senders transmitting to a single receiver. Concast communication is used in a wide range of applications including collaborative applications, report-in style applications, or just end-to-end acknowledgements in a reliable multicast protocol. This paper explores the issues involved in designing concast communication services. We examine various message combination methods including concatenation, compression, and reduction to reduce the traffic loads imposed on the network and packet implosion at the receiver. Group management operations such as group creation/deletion, joining/leaving, and concast routing are discussed. We also address transmission issues such as reliable delivery, flow control, congestion control, and QoS. We conclude the paper by presenting a concast communication model that we have been developing in the context of TMTP5. The model uses concast communication to implement reliable multicast and it shares concast trees with the multicast group whenever possible to reduce overhead costs.
The multiflow conversation protocol (MCP), which provides two communication abstractions is discu... more The multiflow conversation protocol (MCP), which provides two communication abstractions is discussed. MCP provides a token-based mechanism for concurrency control among participants of a multipoint connection and includes a communication abstraction called a multiflow conversation. A conversation may consist of one or more connections, and MCP enforces temporal synchronization in the delivery of traffic over participant connections. Delivery of traffic in a conversation is also based on a notion of causality that takes into account the delay constraints associated with real-time traffic
Raj Yavatkar, Intel Corporation, will focus on the architecture challenges of platform designed w... more Raj Yavatkar, Intel Corporation, will focus on the architecture challenges of platform designed with many core processors – processors with large number of internal cores. Challenges for many-core-based platforms include, but not limited to, power management, reliability and balancing processor-memory-I/O bandwidth requirements. Renato Recio, IBM Corporation, will focus on I/O network evolution and how I/O architectures are addressing the requirements gaps including performance, connectivity and power management. Philip Emma, IBM Corporation, discusses system-on-chip evolutions and the challenges to meet networking bandwidth and storage capacity requirements.
SoC designs are bringing ever increasing functionality to the users fueled by smart devices and n... more SoC designs are bringing ever increasing functionality to the users fueled by smart devices and new user experiences made possible by such devices connected always and everywhere. As more and more functionality is integrated, delivering a complete hw/sw stack and solutions on an aggressive time scale is a challenging task. This talk will discuss where the trends are, what the next wave of innovations and challenges are across many vectors such as systems, system software, CAD, heterogeneity, and programming abstractions.
On behalf of the program committee we are both pleased to welcome you to the third edition of the... more On behalf of the program committee we are both pleased to welcome you to the third edition of the ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS) being held in Orlando, Florida. ANCS has become one of the premier research conferences in networking, and is dedicated to the design of the hardware and software components used to create communication networks. This year, we sought to expand the reach of ANCS to include novel architectures, architectural support for advanced communications systems, algorithms and protocols and emphasize hardware and software elements in the context of systems that enable networks to evolve and scale. We achieved that to a certain extent, and hope that in the future ANCS will be even more successful in addressing system issues for enabling high speed networks operating in large scale. There are many integral components and topics that inform the design of such networks, including switches, routers, optical and wireless networks, packet processing architectures, network processor design, scheduling and security, and this year's program includes papers in all these areas. We were fortunate to recruit an outstanding program committee of 32 members who graciously contributed their time and expertise towards the success of the conference. Every one of them was dedicated and thoroughly reviewed the papers they were assigned, and we are sure the authors will be greatly helped by the feedback provided in the reviews. The quality of any conference is derived from the quality of the submitted papers and the quality of the paper selection process. This conference is indebted to all authors who submitted their best work and to the program committee made of experts in the field. We received 70 submissions from all over the world. About 50% of the submitted papers were from the U.S., and about 28% from Europe, Middle East and Africa, and 20% were from Asia. Every paper received at least three reviews mostly from program committee members. The review process was "double-blind". Authors' identities were kept anonymous throughout the process. At the end the program committee accepted 20 papers which will be presented in a single track over the two days of the conference. We also chose 5 papers from the submitted papers to be presented as posters, with an associated 2 page abstract that is included in the Proceedings. We hope that all of you will find the program enjoyable, stimulating and useful to your career, for this would be the best reward for all the work that went into its preparation.
This paper presents a novel approach to high-performance video processing that avoids the use of ... more This paper presents a novel approach to high-performance video processing that avoids the use of special purpose hardware and achieves parallelism via a distributed system of high- performance workstations. An efficient network protocol allows mobile computers or mobile robots to remotely access the distributed video processing system over low-bandwidth wireless links. Despite the lack of special purpose hardware and expensive computing equipment, the system provides exceptional performance processing data at speeds that approach real-time. The novel contribution of our approach is the ability to process video in compressed format and efficiently distribute video data to a distributed system of workstations for processing. We describe the algorithms employed to achieve these features and demonstrate our approach by showing how it can be applied to two problem domains: multiple target tracking and video archive searching. We present experimental results that show that video data can be efficiently disseminated to a network or workstations with only a small (7%) increase in transfer time. We also show that processing can be performed efficiently (20 fps) directly on the compressed video without social purpose hardware.
... 2. [clouds] Susan Angebranndt, Richard L. hyde, Daphne Huetu Luong, and Nagendra Siravara. In... more ... 2. [clouds] Susan Angebranndt, Richard L. hyde, Daphne Huetu Luong, and Nagendra Siravara. Integrating audio and telephony in a distributed workstion environment. In USENIX '91 Conference Proceedings, pages 419-435. USENIX, June 1991. ... 21. [mcp-dcs] Raj Yavatkar. ...
This paper defines a new group communication model called concast communication. Being the counte... more This paper defines a new group communication model called concast communication. Being the counterpart to multicast, concast involves multiple senders transmitting to a single receiver. Concast communication is used in a wide range of applications including collaborative applications, report-in style applications, or just end-to-end acknowledgements in a reliable multicast protocol. This paper explores the issues involved in designing concast communication services. We examine various message combination methods including concatenation, compression, and reduction to reduce the traffic loads imposed on the network and packet implosion at the receiver. Group management operations such as group creation/deletion, joining/leaving, and concast routing are discussed. We also address transmission issues such as reliable delivery, flow control, congestion control, and QoS. We conclude the paper by presenting a concast communication model that we have been developing in the context of TMTP5. The model uses concast communication to implement reliable multicast and it shares concast trees with the multicast group whenever possible to reduce overhead costs.
The multiflow conversation protocol (MCP), which provides two communication abstractions is discu... more The multiflow conversation protocol (MCP), which provides two communication abstractions is discussed. MCP provides a token-based mechanism for concurrency control among participants of a multipoint connection and includes a communication abstraction called a multiflow conversation. A conversation may consist of one or more connections, and MCP enforces temporal synchronization in the delivery of traffic over participant connections. Delivery of traffic in a conversation is also based on a notion of causality that takes into account the delay constraints associated with real-time traffic
Raj Yavatkar, Intel Corporation, will focus on the architecture challenges of platform designed w... more Raj Yavatkar, Intel Corporation, will focus on the architecture challenges of platform designed with many core processors – processors with large number of internal cores. Challenges for many-core-based platforms include, but not limited to, power management, reliability and balancing processor-memory-I/O bandwidth requirements. Renato Recio, IBM Corporation, will focus on I/O network evolution and how I/O architectures are addressing the requirements gaps including performance, connectivity and power management. Philip Emma, IBM Corporation, discusses system-on-chip evolutions and the challenges to meet networking bandwidth and storage capacity requirements.
SoC designs are bringing ever increasing functionality to the users fueled by smart devices and n... more SoC designs are bringing ever increasing functionality to the users fueled by smart devices and new user experiences made possible by such devices connected always and everywhere. As more and more functionality is integrated, delivering a complete hw/sw stack and solutions on an aggressive time scale is a challenging task. This talk will discuss where the trends are, what the next wave of innovations and challenges are across many vectors such as systems, system software, CAD, heterogeneity, and programming abstractions.
On behalf of the program committee we are both pleased to welcome you to the third edition of the... more On behalf of the program committee we are both pleased to welcome you to the third edition of the ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS) being held in Orlando, Florida. ANCS has become one of the premier research conferences in networking, and is dedicated to the design of the hardware and software components used to create communication networks. This year, we sought to expand the reach of ANCS to include novel architectures, architectural support for advanced communications systems, algorithms and protocols and emphasize hardware and software elements in the context of systems that enable networks to evolve and scale. We achieved that to a certain extent, and hope that in the future ANCS will be even more successful in addressing system issues for enabling high speed networks operating in large scale. There are many integral components and topics that inform the design of such networks, including switches, routers, optical and wireless networks, packet processing architectures, network processor design, scheduling and security, and this year's program includes papers in all these areas. We were fortunate to recruit an outstanding program committee of 32 members who graciously contributed their time and expertise towards the success of the conference. Every one of them was dedicated and thoroughly reviewed the papers they were assigned, and we are sure the authors will be greatly helped by the feedback provided in the reviews. The quality of any conference is derived from the quality of the submitted papers and the quality of the paper selection process. This conference is indebted to all authors who submitted their best work and to the program committee made of experts in the field. We received 70 submissions from all over the world. About 50% of the submitted papers were from the U.S., and about 28% from Europe, Middle East and Africa, and 20% were from Asia. Every paper received at least three reviews mostly from program committee members. The review process was "double-blind". Authors' identities were kept anonymous throughout the process. At the end the program committee accepted 20 papers which will be presented in a single track over the two days of the conference. We also chose 5 papers from the submitted papers to be presented as posters, with an associated 2 page abstract that is included in the Proceedings. We hope that all of you will find the program enjoyable, stimulating and useful to your career, for this would be the best reward for all the work that went into its preparation.
Uploads
Papers by Raj Yavatkar