Computer Network Technology Corporation

Brocade Communications Systems, Inc.

William C Collette

Mark J Moran

Robert Engebretson

Mike Morandi

Steve Flattum

Brian A Johnson

Bret Ketchum

Mark J Vogel

Mark Mansee

Dana K Richard

Roger T Richardson

Lisa D Ludgate

Robert R Beyer

Richard L Cain

Jim Kunz

Richard Kelley

A storage router and related method are presented for combining multiple host frames, such as Fiber Channel frames, together into a single datagram for tunneling transmission over an IP or similar network. The storage router operates by storing incoming host frames in a host frame buffer. When there is sufficient data in the buffer, multiple host frames are batched together, compressed, and converted into an IP datagram. The number of host frames to be batched together can be established through a variety of tests, including total bytes of data, number of frames, or through a time-out mechanism. The network layer then fragments the datagram into data link level frames, such as 1500 byte Ethernet frames. When the datagram arrives at the final destination, the segmented datagram is reconstructed, decompressed, and the multiple host frames are then extracted and passed on to the recipient host frame network.

Batching and compression for IP transmission

A method and apparatus is presented for performing a sequence-level CRC calculation on fiber channel communications within a switching platform domain. A CRC generator searches the data communication for frames that contain the type of data for which a sequence-level CRC is desired, such as for a sequence containing SCSI data. If found, and the type of data allows multiple frames per sequence, the present invention creates a CRC value for the sequence. An intermediate CRC value is stored in a queue to allow the simultaneous calculation of sequence level CRC values for multiple frames. With inbound data, the sequence-level CRC is appended to the end of the sequence data. With outbound data, the calculated value is compared with the appended, expected value, With single-frame fiber channel protocols, the frame-level CRC value is obtained directly from the frames entering the switching platform domain. This value is placed in a local queue, from which it is appended to the data payload by a processor for transmission within the switching platform. When the single frame is leaving the switching platform domain, the flow of data leaving the switching platform domain is monitored and the frame-level CRC value calculated by the fiber channel controller is replaced with the original frame-level CRC value.

Frame-level fibre channel CRC on switching platform

A storage router and related method are presented for combining multiple Fibre Channel frames together into a single IP datagram for tunneling transmission over an Internet protocol network. The storage router operates by storing incoming Fibre Channel frames in a Fibre Channel frame buffer. When there is sufficient data in the buffer, multiple Fibre Channel frames are taken from the buffer and combined into a single IP datagram. The number of Fibre Channel frames to be combined can be established through a variety of tests, including total bytes of data, number of frames, or through a time-out mechanism. The network layer then fragments the datagram into data link level frames, such as 1500 byte Ethernet frames. When the IP datagram arrives at the final destination, the segmented IP datagram is reconstructed, and the multiple Fibre Channel frames are extracted from the IP datagram and passed on to the recipient Fibre Channel network.

Fibre channel frame batching for IP transmission

A storage router and related method are presented for combining multiple host frames, such as Fibre Channel frames, together into a single datagram for tunneling transmission over an IP or similar network. The storage router operates by storing incoming host frames in a host frame buffer. When there is sufficient data in the buffer, multiple host frames are batched together, compressed, and converted into an IP datagram. The number of host frames to be batched together can be established through a variety of tests, including total bytes of data, number of frames, or through a time-out mechanism. The network layer then fragments the datagram into data link level frames, such as 1500 byte Ethernet frames. When the datagram arrives at the final destination, the segmented datagram is reconstructed, decompressed, and the multiple host frames are then extracted and passed on to the recipient host frame network.

Frame batching and compression for IP transmission

The present disclosure is directed to a method and system for analyzing network performance by creating benchmark test results when the computer network is initialized, and subsequently testing the network while operational to determine degradations. This method and system can proactively determine problems before they result in network failures, or immediately detect network failures even before the user is aware of them. In one aspect, the disclosure is directed to a method of monitoring at least one computer network from a remote location. The method includes benchmark-testing at least one computer network with test data provided from the remote location to obtain benchmark test results. Often, benchmark-testing is performed when the network is initialized. Once operational, the remote location will perform operational-testing of at least one computer network with sample data provided from the remote location to obtain operational test result. Operational-testing is performed while at least one computer network is carrying live data storage traffic. The operational test results are compared to the benchmark test results to determine network degradation. Immediate failures can be determined as well as trending of test results can be used to predict failures.

Computer network monitoring with test data analysis

A method and apparatus is presented for performing a sequence-level CRC calculation on fiber channel communications within a switching platform domain. Specifically, the disclosed invention utilizes a CRC generator to monitor data communication between an external interface and a fiber channel controller. The CRC generator searches the data communication for frames that contain the type of data for which a sequence-level CRC is desired, such as for a sequence containing SCSI data. When found, the CRC generator performs an 8B/10B decode, strips off any fill bytes, and performs a CRC calculation on the data payload. If the found frame is the first frame in the sequence, the CRC is done starting with a reset value. If the frame is not the first, an intermediate value containing the previous CRC calculation results for this sequence is retrieved and used to perform the CRC calculations. Upon completion of the calculation for the frame, a determination is made whether the sequence is complete. If so, the calculated value is the sequence-level CRC. If not, the calculated value is stored in memory for use with the next frame in that sequence. With inbound data, the calculated sequence-level CRC is appended to the end of the data comprising that sequence. With outbound data, the calculated value is compared with the appended, expected value, and a CRC error is generated if the values do not match. The sequence-level CRC value appended to the sequence data is typically removed before the data exits the switching platform domain.

Fiber channel CRC for internal error checking on a switching platform

A method and apparatus is presented providing high-performance lossless data compression implemented in hardware for improving network communications. A compression module useful in a switching platform is also presented capable of compressing data stored in buffer memory. Instructions for a compression task are assigned to the compression module by a microprocessor writing a control block to a queue in stored local memory. The control block informs the compression module of the size and location of the unprocessed data, as well as a location in the buffer memory for storing the processed data and the maximum allowed size for the compressed data. Using this technique, the microprocessor can limit the compression of data to those data streams allowing compression, to those segments that are susceptible to compression, and to those segments that are large enough to show a transmission speed improvement via compression.

Growing community of inventors

Stillwater, MN, United States of America

William C Collette