Vice President, Systems Engineering at Arista Networks
- San Francisco Bay Area
- Computer Networking
Doug Gourlay's Overview
Doug Gourlay's Volunteer Experience & Causes
Causes I care about:
- Science and Technology
- Veterans and Military Services
Organizations I support:
- Wounded Warrior Project
- American Red Cross
Doug Gourlay's Publications
From Wall Street to the military: Advances in FPGA switching technology and software-to-hardware development tools facilitate wire-speed data transmissio
- Military Embedded Systems
- June 1, 2012
Low-latency technology "paid for" by Wall Street traders seeking microsecond advantages is trickling up to the military.
Doug Gourlay's Patents
- United States Patent 7,974,196
- Issued July 5, 2011
Methods and systems for using counters to monitor port buffers are described. In one embodiment, the method includes receiving data packets into a system port buffer having different queues. Also, counters are associated with the queues. Further, the method includes scanning the counters, in which the counters count the number of dropped data packets associated with an individual queue of the different queues. Moreover, the method includes determining if the number of dropped data packets for an individual queue is greater than a predetermined number. If the number of dropped data packets is greater than the predetermined number, then a message is generated.
- United States Patent 7,062,560
- Issued December 14, 2004
More effective load balancing by a Server Load Balancer may be achieved by implementing extended Server Load Balancing information in a server. The extended Server Load Balancing information includes a special listing of “hot” objects, or objects that have been designated for special handling by the Server Load Balancer. In order to determine which objects on a server are “hot” at any particular point in time, the available throughput for the server is determined. This is then multiplied by a “hotness” percentage, resulting in an Object Threshold Value (OTV). Each of N objects is then assigned an Object Request Value (ORV), the N objects determined by taking the N objects utilizing the most bandwidth. The ORVs are then compared with the OTV, and any of the objects whose ORV exceeds the OTV are labeled as “hot”. Finally, information on the “hot” objects is communicated to the Server Load Balancer (SLB) for special handling.
- United States Patent 6,850,980
- Issued February 1, 2005
Layer 7 switching may be accomplished using one or more caches placed throughout a computer network. Changes to a file on a server may be detected and propagated throughout the network. At the switch or router level, once notification of changes to a file is received, the content may be retrieved from the server and placed in a connected cache. A routing table entry may be created for the content and also placed in the cache. The routing table entry may contain an original location field identifying the original location of the content, a distance field indicating a distance from the cache to the server, and a field indicating a version number of the content. Additional fields may also be contained within the routing table entry. When a user requests a specific file, rather than forward the request directly to the server containing the original file, the request may be handled by the router closest to the user which has a connected cache containing the content. This allows a user's...
Inventors: Doug Gourlay, Ajit Sanzgiri
- United States Patent 7,616,613
- Issued November 10, 2009
Enhanced tunnel communication mode creation, management and tuning in a network that includes wireless access points (APs) and user authentication. Tunnels can be dynamically managed to adapt to the changing topology of a network with APs. User devices such as mobile phones, laptop computers, personal digital assistants, or other devices can be added or dropped from an assigned AP. APs, routers, switches or other devices can also be added, removed, or modified in their network characteristics. Special control is also provided for IP multicast, Dynamic Host Configuration Protocol (DHCP), Address Resolution Protocol (ARP) and other network features.
- United States Patent 6,820,123
- Issued November 16, 2004
More effective load balancing by a Server Load Balancer may be achieved by implementing extended Server Load Balancing information in a server. The extended Server Load Balancing information includes a special listing of "hot" objects, or objects that have been designated for special handling by the Server Load Balancer. In order to determine which objects on a server are "hot" at any particular point in time, the available throughput for the server is determined. This is then multiplied by a "hotness" percentage, resulting in an Object Threshold Value (OTV). Each of N objects is then assigned an Object Request Value (ORV), the N objects determined by taking the N objects utilizing the most bandwidth. The ORVs are then compared with the OTV, and any of the objects whose ORV exceeds the OTV are labeled as "hot". Finally, information on the "hot" objects is communicated to the Server Load Balancer (SLB) for special handling.
- United States Patent 7,720,997
- Issued May 18, 2010
Methods and apparatus for processing a service request received from a client are disclosed. When a service request is received, a plurality of packets are transmitted, where each of the plurality of packets identify a different type of service via which to send the corresponding packet. In addition, a mapping of each different type of service to an IP address is preferably maintained, thereby enabling the service request to be processed via an IP address associated with a type of service identified in a first one of the plurality of packets to be received.
- United States Patent 7,719,992
- Issued May 18, 2010
A method for cable diagnostics in a network includes performing a test to determine initial state information for each of a plurality of lines coupled to a switch and storing the initial state information in a database. When a change in the state of a line is detected, the test is re-run to determine new state information of the line. The new state information is stored in the database and a message that identifies the change in state and a likely cause of the state change is issued to a network operator. It is emphasized that this abstract is provided to comply with the rules requiring an abstract that will allow a searcher or other reader to quickly ascertain the subject matter of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.
- United States Patent 7,516,211
- Issued April 7, 2009
A data communication device automatically configures its own communication ports based on attributes of a remote device with which it communicates. The process of configuring the communication ports includes monitoring a communications protocol associated with a remote device on a given communication port of the data communication device without participating in the communications protocol. The data communication device detects an attribute such as a network address associated with the remote device based on the monitored communications. In response to detecting the attribute of the remote device, the data communication device selects one of multiple configuration profiles depending on the detected attribute of the remote device to configure the given communication port.
Inventors: Doug Gourlay, James Aviani, Carl Sutton
- United States Patent 7,624,184
- Issued November 24, 2009
The invention is directed to techniques for managing access to data available from a data access device using a network switching protocol to manage content requests from clients. A data communications device receives a request for content from a client. The data communications device, such as a network switch, exchanges signals using the network switching protocol with one or more data access devices, such as a web cache. The data communications device can then select one data access device to provide the requested content to the client based, for example, on load balancing considerations. The data communications device sends a data transfer approval signal to the selected data access device. The selected data access device establishes a connection with the client using information, such as the client's network address, provided by the signals. The selected data access device then bypasses the data communications device and sends the requested content to the client. Using the..
Inventors: Doug Gourlay, James Aviani
- United States Patent 7,349,969
- Issued March 25, 2008
In order to direct content requests to an appropriate content serving site in a computer network, a phased learning approach is utilized to ensure that, as best as possible, the request is made to content serving site with the shortest delay. In a setup phase, an indirect path return geographic sever load balancer times queries to all of the individual content serving sites so that the queries all arrive at the content serving sites at the same time. Therefore, when the requesting fixed location receives communications from the content serving sites, it can easily tell which content serving site has the least delay by an established race condition. The winner of the race may then be relayed to the indirect path return geographic server load balancer for later usage. In an execution mode, only the m fastest content serving sites and n other sites (used to test random and new sites) are queried when a DNS request arrives from the requesting fixed location. The particular m fastest...
- United States Patent 7,889,733
- Issued February 15, 2011
An adjunct network device includes several ports, an uplink interface, and an adjunct forwarding engine coupled to the ports and the uplink interface. A first port is configured to receive a packet, which includes a destination address. The adjunct forwarding engine is configured to send the packet to the uplink interface if the destination address is not associated with any of the ports. The packet is sent to one of the ports if the destination address is associated with the one of the ports.
Energy efficient scaling of network appliance service performance
- United States Patent 20110069613
- United States Patent 20080304472
- Issued June 5, 2007
In one embodiment, a source device (e.g., a VOIP phone) establishes a call connection with a remote device depending on which of multiple network paths provides an acceptable latency (e.g., a lower latency). For example, in response to receiving a request to establish a connection with a remote destination device over a network, the source device (e.g., a caller's phone) obtains multiple service code values. The source device encodes each of multiple data packets to include a unique service code value for transmission of the messages over different network topologies to a remote destination. Thus, when transmitted, each of the multiple messages follows a different logical network topology of a network as specified by a respective service code value. Based on feedback from a remote device that receives the multiple messages, the source learns a preferred logical network topology of the network for establishing the call connection.
Network Accessibility to any Network Attached Device During Reboot and Power Loss
- United States Patent 8,509,097
- Issued December 5, 2013
A data communication network (DCN) having a plurality of network devices coupled to the DCN with at least one of the network devices having a "boot once" connectivity manager processor (CMP). The CMP receives its power over the DCN rather than from the power applied to the network devices. The CMP can execute special operating system code and maintain network connectivity even if the network device itself is powered off, is being booted or is otherwise non-functional. The CMP is also coupled to the network device's memory so that it may respond to out-of-band polling requests for device status information from network management tools. With CMP, network administrators can monitor the boot process of network devices, determine that a network device is non-functional due to power loss and can maintain an accurate inventory status of spare network devices that are stored un-powered in a spares closet.
Graceful Removal and/or Insertion of a Network Element
- United States Patent Application 20100002577
- Filed July 2, 2008
In one embodiment, when a network element is to be removed from or inserted into a network a Graceful Operations Manager schedules graceful shut-down and/or start-up routines for different protocols and/or components on the network element in an optimal order based on dependencies between the different protocols and components. The Graceful Operations Manager communicates with the different components at different stages of their shut-down or start-up process and communicates information on the standby topology across components and/or protocols to enable the synchronization of the standby topology computation on all components and/or protocols that are affected by the removal or insertion.
System and method for selectively affecting a computing environment based on sensed data
- United States Patent Application 20070260417
- Filed November 8, 2007
A system and method for affecting computing resources. The method includes sensing variables associated with spatially dispersed computing resources and providing sensed data in response thereto. Subsequently the spatially dispersed computing resources are selectively automatically affected based on sensed variables associated with the computing resources. In a specific embodiment, the method further includes determining if the sensed data meet a predetermined criterion or criteria and providing one or more control signals in response thereto. The specific method further includes moving virtual machines associated with computing resources that meet the predetermined criterion or criteria to computing resources that do not meet the predetermined criterion or criteria. The sensed data may include temperature, and the predetermined criteria or criterion may include a predetermined threshold beyond which temperature data is considered to meet the predetermined criterion. In an illustrative embodiment, the method further includes selectively activating one or more devices, such as cooling systems, that are adapted to alter sensed variables to cause the sensed data to no longer meet the predetermined criterion or criteria.
System and Method for Device Failure Notification
- United States Patent Application 20140280792
- Filed May 13, 2013
A method and apparatus of a device that notifies another device of a failed device is described. In an exemplary embodiment, a network element detects that a first device is unavailable, where the network element couples the first device to the second device. In response to detecting that the first device is unavailable, the network element configures a proxy for the first device. The network element additionally receives network data that is destined for the first device, where the second device originated the network data. If the proxy can process the network data, the network element transmits a response to the second device from the proxy, where the response indicates that the first device is unavailable, where the first response includes an address of the first device. If the proxy cannot process the network data, the network element drops the network data.
Doug Gourlay's Skills & Expertise
- Data Center
- Load Balancing
- Product Marketing
- Product Management
- Storage Area Networks
- Cloud Computing
- Go-to-market Strategy
- Enterprise Software
- Channel Partners
- Cisco Technologies
- Sales Enablement
- Managed Services
- Unified Communications
- Business Alliances
- Strategic Partnerships
- Professional Services
- Network Architecture
- Network Security
- Solution Selling
- Cisco IOS
- Demand Generation
- Direct Sales
- Lead Generation
- Partner Management
- Network Design
- Sales Engineering
- Datacenter Virtualization
- Financial Markets
- Service Providers
Doug Gourlay's Additional Information