Browsing by Author "Wang, PC"

Now showing 1 - 20 of 20

A fast IP routing lookup scheme
(2000)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
A major design issue for the next generation routers is the 1P lookup mechanism. The router needs to perform a longest prefix matching on the address lookup for each incoming packet to determine the next hop. Currently, the process is done in software and has become a major performance bottleneck of the router. In this paper, we propose a fast IP lookup mechanism in which the forwarding table is small enough to fit in SRAM with very low cost. It also can be implemented in hardware using the pipeline technique. A large routing table with 45,000 routing prefixes can be compressed to a forwarding table with about 430 Kbytes in size by using our proposed method. In the worst case, the number of memory accesses for a lookup is three. When implemented in pipeline technique, the proposed mechanism can achieve one routing lookup per memory access. With current ions SRAM, this mechanism provides approximately 100 million routing lookups per second. Futhermore, the lookup speed can be improved linearly through the speedup of the memory access.
A fast IP routing lookup scheme
(2001-03-01)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
A major issue in router design for the next generation Internet is the fast IP address lookup mechanism. The existing scheme by Huang et al. performs the IP address lookup in hardware in which the forwarding table can be compressed to fit into reasonable-size SRAM, and a lookup can be accomplished in three memory accesses. In this letter, we claim that with a little extra memory, it is able to further reduce the lookup time to two memory accesses.
A fast packet classification by using enhanced tuple pruning
(2002)
Wang, PC
;
Chan, CT
;
Tseng, WC
;
Chen, YC
;
資訊工程學系
In the packet classification, the route and resources allocated to a packet are determined by the destination address as well as other header fields of the packet such as source/destination address, TCP and UDP port numbers. It has been demonstrated that performing packet classification on a potentially large number of fields is difficult and has poor worst-case performance. In this work, we proposed an enhanced tuple pruning search algorithm called "Tuple Pruning +" that provides fast two-dimension packet classification. With reasonable extra filters added for Information Marker, only one hash access to the tuples is required. Through experiments, about 8 MB memory is required for 100K-filter database and 20 million packet per second (MPPS) is achievable. The results demonstrate that the proposed algorithm is suitable for high-speed packet classification.
A fast table update scheme for high-performance IP forwarding
(2002-01-01)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
In the previous work, Lampson et al. proposed an IP lookup algorithm which performs binary search on prefixes (BSP) [1]. The algorithm is attractive, even for IPv6, because of its bounded worst-case memory requirement. To achieve fast forwarding, it may need to slow down the insertion speed. Although this can be justified, the routing-table reconstruction in BSP is too time-consuming to handle the frequent route updates. In this work, we propose a fast forwarding-table construction algorithm which can accomplish more than 4,000 route updates per second. Moreover, it is simple enough to fulfill the need of fast packet forwarding. With the enhanced multiway search tree, we further reduced the depth of the tree and eliminated the pointer storage; this reduces the forwarding table size and shortens the lookup time.
A fast table update scheme for high-performance IP forwarding
(2001)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
There has been an extensive study in constructing the routing tables during the past few years. Although the existing works have certain advantages, ho,vel er, those approaches either use complicated data structures which result in large storage requirement and high complexity updating/building the forwarding table, or they are not scalable refit in IPv6. Lampson et al. proposed an IP lookup algorithm which performs binary search on prefixes (BSP) [11] The algorithm is attractive, even for IPv6, because of its bounded worst-case memory requirement. For achieving fast forwarding the cost paid is the slowing down insertion. Although this can be justified, the performance of routing-table reconstruction in BGP is too time-consuming,ring to handle the frequent route updates. In this work, we propose a fast forwarding-table construction algorithm which can handle more than 4,000 route updates per second. Moreover; it is simple enough to fufil the need of fast packet forwarding. With the modified multiway search tree, we can further reduce the depth of the tree and eliminate the storage for pointers, this reduces the fowarding table size and shorten the lookup time.
A FIFO-based buffer management approach for the ATM GFR services
(2000-06-01)
Chan, CT
;
Hu, SC
;
Wang, PC
;
Chen, YC
;
資訊工程學系
We investigate the TCP packet transmission over the ATM Guaranteed Frame Rate (GFR) service using a selective packet-discard with buffer vacancy tracking strategy and a packet-based push-out buffering scheme. Our approach is based on FIFO queueing discipline, it features fair sharing of available resources as well as feasibility in hardware implementation.
An efficient IP routing lookup by using routing interval
(2001-12-01)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP packets forwarding, this complicates the IP lookup. Currently, this process is mainly performed in software and several schemes have been proposed for hardware implementation. Since the IP lookup performance is a major design issue for the new generation routers, in this article we investigate the properties of the routing table and present a new approach for IP lookups, our approach is not based on BMP and significantly reduces the complexity, thus the computation cost of existing schemes can be significantly reduced. We also propose an efficient IP lookup algorithm, with which we improve the binary search on prefixes to 30 millions of packets per second (MPPS) and 5,000 route updates/s under the same experiment setup with an even larger routing table.
An efficient traffic control approach for GFR services in IP/ATM internetworks
(1998)
Chan, CT
;
Chen, YC
;
Wang, PC
;
資訊工程學系
;
Department of Computer Science
The Guaranteed Frame Rate (GFR)/UBR+ service has been proposed to support traffic streams with a minimum packet level guarantee for packets not exceeding the maximum size. In addition, the service also specifies that the excess traffic of a user should have access to a fair share of available resource. In this paper, we investigate a packet-discard strategy in combination with the push-out scheme, which alleviates the effect of fragmentation and supports GFR services efficiently. The simulation results show that our proposed method achieves the requirements of GFR service as well as improves the TCP throughput using the common FIFO cell scheduling. Moreover, it is feasible in implementation.
An efficient traffic control scheme for TCP over ATM GFR services
(2002-04-05)
Chan, CT
;
Wang, PC
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
In ATM networks, the guaranteed frame rate (GFR) service category has been defined to support user applications which are neither able to specify the range of traffic parameter values, nor able to comply with the behavior rules. It provides a packet-level rate guarantee with a given maximum packet size. The service specifies that the excessive traffic from a user should share the available resource fairly. In this paper, we investigate TCP/IP packet transmission over ATM by using a selective packet-discard strategy with tracking of the available buffer space and a packet push-out buffering scheme to accommodate the GFR service. The simulation results show that our proposed method fulfills the requirements of GFR service as well as improves the TCP throughput under the common FIFO scheduling. A feasible implementation approach is also addressed. (C) 2001 Elsevier Science B.V. All rights reserved.
An intelligent buffer management approach for GFR services in IP/ATM internetworks
(2000-08-30)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
In ATM networks, the Guaranteed Frame Rate (GFR) service has been proposed to accommodate non-real-time traffic with packet size not exceeding a maximum length, such as TCP/IP based traffic. The GFR service not only guarantees a minimum throughput at the frame level, but also supports a fair share of available resources. In this article, we propose a buffering strategy called Intelligent Fair Buffer Allocation (IFBA) that provides minimum cell rate (MCR) guarantees and fair sharing to GFR virtual circuits (VCs). From the simulation results, we demonstrate that IFBA fulfills the requirements of GFR service as well as improves the TCP throughput using the FIFO scheduling. (C) 2000 Elsevier Science B.V. All rights reserved.
Cost-utility analysis of tympanomastoidectomy for adults with chronic suppurative otitis media
(2005-09-01)
Wang, PC
;
Jang, CH
;
Shu, YH
;
Tai, CJ
;
Chu, KT
OBJECTIVE: To undertake cost-utility analysis for tympanomastoid surgery to analyze itscost-effectiveness in treating adult chronic suppurative otitis media (CSOM). METHODS: Seventy-seven patients with CSOM were evaluated with the Chronic Ear Survey (CES) before and 1 year after tympanomastoid surgery. Direct health care cost data during the 1st year after operation were retrieved. The utility gain was defined as change in the CES total score. The cost-utility ratio (CUR) was defined as cost per utility gain. Patients were stratified by disease type into wet-ear and dry-ear groups. RESULTS: The average total direct cost attributable to tympanomastoid surgery is (in New Taiwan dollars [NT$]) 45,716.3 in the 1st postoperative year, and the average CUR is NT$ 1850.9. The lower CUR of NT$ 1280.9 for the wet-ear group is due to the greater utility gain (37.6 +/- 3.4 versus 24.4 +/- 6.8, P < 0.05) despite its higher cost (NT$ 48,163.2 versus NT$ 38,419.7, P < 0.05). CONCLUSIONS: Treating continuously or intermittently draining ears is more cost-effective, as compared with managing a quiescent infection, because of its favorable utility gain. (C) 2005 American Academy of Otolaryngology.
Design and analysis of traffic control for guaranteed services on the Internet
(2000-05-01)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
Real-time communications services will be one of the most promising applications on the Internet. Real-time traffic usually utilizes a significant amount of resources while traversing through the network. To provide guaranteed service for delay-sensitive real-time applications, both resource reservation and effective traffic control are necessary. The resource reservation protocol (RSVP) was developed to deal with the former problem while various approaches have been proposed to solve the latter one. It is crucial for guaranteed services to maintain satisfactory performance through effective traffic control, which must be implemented in a feasible way and be able to work with RSVP. In this article, we present a flexible and effective traffic-control approach with a resource reservation scheme to accommodate delay-sensitive services on the Internet. We demonstrate that our approach features proven bounded end-to-end queueing delay for delay-sensitive applications and good resilience for resource allocation. In addition, our packet transmission discipline achieves significant improvement in end-to-end delay control.
High-performance IP routing table lookup
(2002-02-15)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP packets forwarding, which complicates the IP lookup. Currently, this process is mainly done in software and several schemes have been proposed for hardware implementation. Since the IP lookup performance is a major design issue for the stew generation routers, in this article we propose a fast IP address lookup scheme, which significantly reduces the forwarding table to fit into SRAM with very low cost. It can also be implemented in hardware using the pipeline technique. By using our proposed method, the required memory space can be 90-170 Kb less than the previous scheme and three memory accesses for a lookup in the worst case. When implemented in hardware pipeline architecture, our mechanism can achieve one routing lookup per memory access. (C) 2002 Elsevier Science B.V. All rights reserved.
Performance enhancement of IP forwarding by reducing routing table construction time
(2001-05-01)
Wang, PC
;
Chan, CT
;
Chen, YC
;
資訊工程學系
;
Department of Computer Science
In previous work, Lampson et al, proposed an IP lookup algorithm which performs binary search on prefixes (BSP) [3], The algorithm is attractive for IPv6 because of its bounded worst-case memory requirement. Although for the sake of fast forwarding, the cost paid for the slowing down insertion is reasonable, the performance of routing-table reconstruction in BGP is too time-consuming to handle the frequent route updates. In this letter, we propose a fast forwarding-table construction algorithm which can handle more than 3600 route updates per second, Moreover, it is simple enough to fulfill the need of fast packet forwarding.
Performance enhancement of XCP for high bandwidth-delay product networks
(2005)
Yang, YR
;
Chan, CT
;
Wang, PC
;
Chen, YC
;
資訊工程學系
It has been demonstrated that as the per-flow product of bandwidth delay increases, TCP becomes inefficient and unstable, regardless of the queuing discipline. By decoupling utilization control from fairness control, the eXplicit Control Protocol (XCP) outperforms TCP in both conventional and high bandwidth-delay environments. In this article,. we present an efficient flow control scheme based on XCP for high bandwidth-delay product networks. It can shorten the response time and improve the performance of XCP. Simulation results show the effectiveness of our proposed scheme in several performance aspects. Moreover, its implementation is simple enough to fulfill the required criteria of high bandwidth-delay product networks.
Performance improvement of hardware-based packet classification algorithm
(2005)
Chen, YC
;
Wang, PC
;
Lee, CL
;
Chan, CT
;
資訊工程學系
Packet classification is important in fulfilling the requirements of differentiated services in next generation networks. In the previous work, we presented an efficient hardware scheme, Condensate Bit Vector, based on bit vectors. The scheme significantly improves the scalability of packet classification. In this work, the characteristics of Condensate Bit Vector are further illustrated, and two drawbacks that may negatively affect the performance of Condensate Bit Vector are revealed. We show the solution to resolve the weaknesses and introduce the new schemes, Condensate Ordered Bit Vector and Condensate and Aggregate Ordered Bit Vector. Experiments show that our new algorithms drastically improve the search speed as compared to the original algorithm.
Routing interval: A new concept for IP lookups
(2001)
Wang, PC
;
Chan, CT
;
Hu, SC
;
Chen, YC
;
資訊工程學系
Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP packets forwarding, which complicates the IP lookup. Currently, this process is mainly done in software and several schemes have been proposed for hardware implementation. Since the IP lookup performance is a major design issue for the new generation routers, in this article we investigate the properties of the routing table and present a new approach for EP lookups, our approach does not require the rule of BMP and thus significantly reduces the complexity. By applying our proposed approach, the computation cost of existing schemes can be significantly reduced. We also propose an efficient IP lookup algorithm, with which we improve the binary search on prefixes to 30 MPPS and 5,000 route updates/s under the same experiment setup with an even larger routing table.
Thermal stability of selective chemical vapor deposited tungsten contact and effects of in situ N-2 plasma treatment
(1998-07-01)
Wang, MT
;
Wang, PC
;
Chuang, MC
;
Chen, LJ
;
Chen, MC
This work investigates the thermal stability of Al/W/p(+)-n junction diodes, in which the W contact was filled using selective chemical vapor deposition to a thickness of about 450 nm and served as diffusion barrier between the Al and the Si substrate. The effects of in situ N-2 plasma treatment on the barrier effectiveness were also investigated. The Al/W(450 nm)/p(+)-n junction diodes can sustain a 30 min furnace annealing up to 575 degrees C. With an in situ N-2 plasma treatment on the W surface caused a thin layer of WNx to form on the W surface, and the nitrified layer of WNx/W acting as barrier between the AZ and the Si substrate effectively suppressed WAl12 formation at elevated temperatures, resulting in a significant barrier improvement. N2 plasma treatment at 100 W for 300 s enabled the Al/WNx/W(450 nm)lp(+)-n junction diodes to sustain thermal annealing at temperatures up to 625 degrees C without degradation of electrical characteristics. (C) 1998 American Vacuum Society.
Traffic control approaches for voice over ATM networks
(1998-12-01)
Chen, YC
;
Chan, CT
;
Hu, SC
;
Wang, PC
;
資訊工程學系
In this paper we present two traffic control approaches, a circuit emulation traffic control (CETC) and an adaptive priority traffic control (APTC) for supporting voice services in ATM networks. Most voice services can be handled as CBR traffic, this causes a lot of wasted bandwidth. Sending Voice through VBR(variable bit rate) may be a better alternative. because it allows the network to allocate voice bandwidth on demand. In CETC. the service discipline guarantees the quality of service (QOS) for voice circuits. Through mathematical analysis. we show that CETC features an adequate performance in delay-jitter. Moreover, it is feasible in implementation. We also present an APTC approach which uses a dynamic buffer allocation scheme to adjust the buffer size based on the real traffic need, as well as employs an adaptive priority queuing technique to handle various delay requirements for VER voice traffic. It provides an adequate QOS for voice circuits in addition to improving the multiplexing gain. Simulation results show that voice traffic get satisfied delay performance using our approaches. It may fulfill the emerging needs of voice service over ATM networks.
Traffic control approaches for voice over ATM networks
(1998-12-01)
Chen, YC
;
Chan, CT
;
Hu, SC
;
Wang, PC
;
資訊工程學系
In this paper we present two traffic control approaches, a circuit emulation traffic control (CETC) and an adaptive priority traffic control (APTC) for supporting voice services in ATM networks. Most voice services can be handled as CBR traffic, this causes a lot of wasted bandwidth. Sending Voice through VBR(variable bit rate) may be a better alternative. because it allows the network to allocate voice bandwidth on demand. In CETC. the service discipline guarantees the quality of service (QOS) for voice circuits. Through mathematical analysis. we show that CETC features an adequate performance in delay-jitter. Moreover, it is feasible in implementation. We also present an APTC approach which uses a dynamic buffer allocation scheme to adjust the buffer size based on the real traffic need, as well as employs an adaptive priority queuing technique to handle various delay requirements for VER voice traffic. It provides an adequate QOS for voice circuits in addition to improving the multiplexing gain. Simulation results show that voice traffic get satisfied delay performance using our approaches. It may fulfill the emerging needs of voice service over ATM networks.