Robert Alan Williams

Bio: Robert Alan Williams is an academic researcher from Advanced Micro Devices. The author has contributed to research in topics: Network packet & Network interface. The author has an hindex of 21, co-authored 57 publications receiving 1439 citations. Previous affiliations of Robert Alan Williams include GlobalFoundries.

Method and apparatus for performing priority-based flow control

Abstract: A network device that controls the communication of data frames between stations receives data frames having different levels of priority. The network device identifies the levels of priority and processes the frames based on the priority level. When a congestion condition associated with a resource on the network device occurs, the network device generates a pause frame that includes a priority indicator and transmits the pause frame to at least one station. When a receiving station receives the pause frame, the receiving station suspends transmission of data frames having a priority corresponding to the priority indicator and continues transmitting frames having other priorities.

Apparatus and method for selectively controlling transmission of consecutive packets in a network station

Abstract: Delay times are modified in an Ethernet network device having captured the media channel by increasing the interframe spacing (IFS) between data packets. The modified IFS interval, increased by adding a delay interval to the minimum interpacket gap (IPG) interval after a first user-selectable number of consecutive successful transmissions, enables other network stations to transmit data during the deferral interval. The Ethernet network device maintains the modified IFS for a limited deferral interval, based upon a predetermined time interval, a number of successful transmissions by other network stations, or a second user-selectable number of consecutive successful transmissions. Additional delay intervals may be added if the network station continues to exceed the predetermined number of consecutive successful transmissions. Burst transmission can also be optimized by counting the consecutive successful transmissions only after detecting an attempted transmission by another network station. Hence, a network station can avoid capturing a network channel while ensuring access latencies.

Method and apparatus for controlling the flow of data frames through a network switch on a port-by-port basis

Abstract: An apparatus and method are disclosed for regulating the flow of data between plural network stations through a network switch. A receive port functions to receive data frames from a first network station, and a transmit port outputs the received data frames to a second network station. A programmable threshold register is provided for storing a threshold value that indicates a saturation level for the internal resources of the transmit port. Control circuitry is used to monitor the internal resources of the transmit port and determine whether or not the threshold value has been reached. If the threshold value has been reached, then the control circuitry will implement a flow control process that causes the first network station to discontinue transmission of data frames to the transmit port until the internal resources of the transmit port fall below the threshold value.

Apparatus and method in a network interface for enabling power up of a host computer using magic packet and on-now power up management schemes

Abstract: A network interface for a workstation having multiple power supply domains includes a wake up module for detecting a wake up request in a received data packet according to the on-now power management scheme and Magic Packet™ power management schemes. An EEPROM supplies an override bit to ensure that the on-now power management schemes and Magic Packet™ power up management schemes can have co-existent detection mechanisms, independent of whether a required enable bit is set by the operating system. The disclosed arrangement maintains the power management schemes in the event that a power loss disables the enabled bit normally supplied by the host computer operating system.

Network interface with security association data prefetch for high speed offloaded security processing

Abstract: One aspect of the invention relates to a network interface system for interfacing a host system with a network. The network interface system includes a bus interface system, a media access control system, and a security system. The security system selectively perform security processing on data incoming from the network based on security associations stored in a memory external to the network interface system, typically a host system memory. The security association for any given frame, when available, is fetched from the external memory after the frame begins to arrive in the network interface system based in part on information contained in the frame. Preferably, the fetch begins before the frame is fully received and the security association is queued whereby security processing can begin without having to wait for the security association to be fetched.

Tiered contention multiple access (TCMA): a method for priority-based shared channel access

Abstract: Quality of Service (QoS) support is provided by means of a Tiered Contention Multiple Access (TCMA) distributed medium access protocol that schedules transmission of different types of traffic based on their service quality specifications. In one embodiment, a wireless station is supplied with data from a source having a lower QoS priority QoS(A), such as file transfer data. Another wireless station is supplied with data from a source having a higher QoS priority QoS(B), such as voice and video data. Each wireless station can determine the urgency class of its pending packets according to a scheduling algorithm. For example file transfer data is assigned lower urgency class and voice and video data is assigned higher urgency class. There are several urgency classes which indicate the desired ordering. Pending packets in a given urgency class are transmitted before transmitting packets of a lower urgency class by relying on class-differentiated urgency arbitration times (UATs), which are the idle time intervals required before the random backoff counter is decreased. In another embodiment packets are reclassified in real time with a scheduling algorithm that adjusts the class assigned to packets based on observed performance parameters and according to negotiated QoS-based requirements. Further, for packets assigned the same arbitration time, additional differentiation into more urgency classes is achieved in terms of the contention resolution mechanism employed, thus yielding hybrid packet prioritization methods. An Enhanced DCF Parameter Set is contained in a control packet sent by the AP to the associated stations, which contains class differentiated parameter values necessary to support the TCMA. These parameters can be changed based on different algorithms to support call admission and flow control functions and to meet the requirements of service level agreements.

System and method for routing data on a packet network

Abstract: The disclosed embodiments include system and method for routing communications over a communications network. In one embodiment, a determination is made as to whether to establish a communication along a communication path based on network performance information. The disclosed embodiments provide several alternatives for establishing the communication and improving communications along the communication path in response to a determination that at least one network node along the communication path is impaired or congested.

System and method for source IP anti-spoofing security

Abstract: A system and method that provides for using source IP addresses and MAC addresses in a network to provide security against attempts by users of the network to use false source IP addresses in data packets. The system and method provide for analyzing MAC addresses and source IP addresses at the datalink (layer 2) level, and to use the information derived from such analysis to block access through a port where a host device is using a false, or spoofed, source IP address in transmitted data packets. Further, the system and method provide for validating initially learned source IP addresses, and for determining whether the number of unsuccessful attempts to validate new source IP addresses exceeds a threshold level, and where the number does exceed the threshold number the system and method can provide for operation in a possible attack mode.

Systems and Methods for Providing Dynamic Network Authorization, Authentication and Accounting

Abstract: Systems and methods for selectably controlling and customizing source access to a network, where the source is associated with a source computer, and wherein the source computer has transparent access to the network via a gateway device and no configuration software need be installed on the source computer to access the network. A user may be prevented access from a particular destination or site based upon the user's authorization while being permitted to access to other sites that the method and system deems accessible. The method and system can identify a source without that source's knowledge, and can access customizable access rights corresponding to that source in a source profile database. The source profile database can be a remote authentication dial-in user service (RADIUS) or a lightweight directory access protocol (LDAP) database. The method and system use source profiles within the source profile database to dynamically authorize source access to networks and destinations via networks.

Systems and methods for providing content and services on a network system

Abstract: Systems and methods for managing and providing content and services on a network system. Aspects of the invention include location-based determination of network content and services that may be provided to client computers. Other aspects of the invention include authorization and authentication components that determine access rights of client computers. Additional aspects include systems and methods for redirecting client computers to different network content. The disclosed systems and methods may be used in numerous network system applications.