Robert Franklin has over 20 years of experience as a senior network engineer and systems administrator. He is currently a senior network engineer at Shutterstock where he leads the design and operation of their network including datacenters, security, and performance monitoring. Previously he held senior roles at Catalyst Systems Repository, WildBlue Communications/ViaSat, and other startups where he designed and implemented large ISP and datacenter networks.
VL2: A scalable and flexible Data Center NetworkAnkita Mahajan
This Data Center Network Architecture introduces a virtual layer 2.5 in the protocol stack of hosts and uses a directory service to achieve efficient forwarding. It uses separate location/identifier IPs
VL2: A scalable and flexible Data Center NetworkAnkita Mahajan
This Data Center Network Architecture introduces a virtual layer 2.5 in the protocol stack of hosts and uses a directory service to achieve efficient forwarding. It uses separate location/identifier IPs
Presentation from October 2012 RTI Technical Road Show.
Agenda Highlights:
How the DDS standard fosters information sharing and interoperability across systems of systems while driving down development, integration, maintenance, upgrade and acquisition costs
The latest 5.0 release of RTI's DDS solution and future roadmap, including enhanced security, support for integration patterns common in C2 systems, FAA DO-178C Level A certification, and DDS standardization initiatives
RTI's new Open Community Source license, which provides free-of-charge access to RTI DDS and allows it to be freely shared across projects and organizations
Layer 3 provides services to exchange the individual pieces of data over the network between identified end devices.
Understanding L3 addressing, encapsulation, routing and decapsulation.
Routed protocols, routing protocols and other supporting protocols such as ICMP and ARP.
IP v4 Characteristics:
1. Connectionless
2. Media independent
3. Best effort
IP v4 Header:
1. Packet length
2. Type of Service
3. Time to Live
4. Protocol
5. Header checksum
6. Source address
7. Destination address
8. Fragment offset.
9. IHL
10. Type of Service
This document contains study of Peer to Peer Distributed system.Three Models of Distributed system.Such as Centralizes,Decentralized,Hybird Model and Pros and cons of these models. Skpye and Bit torrent architecture is also discussed.This tutorial can be very help full for those who are beginners.
IEEE - Standards.
Ethernet Logical Link Control (LLC) and Media Access Control (MAC) sublayers.
Historic Ethernet: The foundation for Ethernet technology in 1970 was Alohanet.
Early Ethernet Media used coaxial cable to connect computers in a bus topology. It was known as Thicknet (10BASE5) and Thinnet (10BASE2).
Ethernet Collision Management.
There are two styles of Ethernet framing: IEEE 802.3 (original) and the revised IEEE 802.3 (Ethernet).
CSMA/CD, Carrier Sensle, Multiple Access, Collision Detection, JAM Signal and Random Backoff.
Ethernet Timing and latency.
Bit Time: The period of time required for a bit to be placed and sensed on the media.
Slot Time: is the time it takes for an electronic pulse to travel the length of the maximum theoretical distance between two nodes.
Inter-frame spacing.
Ethernet Types
"OpenHPC is a collaborative, community effort that initiated from a desire to aggregate a number of common ingredients required to deploy and manage High Performance Computing (HPC) Linux clusters including provisioning tools, resource management, I/O clients, development tools, and a variety of scientific libraries. Packages provided by OpenHPC have been pre-built with HPC integration in mind with a goal to provide re-usable building blocks for the HPC community. Over time, the community also plans to identify and develop abstraction interfaces between key components to further enhance modularity and interchangeability. The community includes representation from a variety of sources including software vendors, equipment manufacturers, research institutions, supercomputing sites, and others."
Watch the video: http://wp.me/p3RLHQ-gKz
Learn more: http://openhpc.community/
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
Trends in Systems and How to Get Efficient Performanceinside-BigData.com
In this video from Switzerland HPC Conference, Martin Hilgeman from Dell presents: HPC Workload Efficiency and the Challenges for System Builders.
"With all the advances in massively parallel and multi-core computing with CPUs and accelerators it is often overlooked whether the computational work is being done in an efficient manner. This efficiency is largely being determined at the application level and therefore puts the responsibility of sustaining a certain performance trajectory into the hands of the user. It is observed that the adoption rate of new hardware capabilities is decreasing and lead to a feeling of diminishing returns. This presentation shows the well-known laws of parallel performance from the perspective of a system builder. It also covers through the use of real case studies, examples of how to program for energy efficient parallel application performance."
Watch the video: http://wp.me/p3RLHQ-gIS
Learn more: http://dell.com
and
http://www.hpcadvisorycouncil.com/events/2017/swiss-workshop/agenda.php
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
Switching: means receiving data on a certain port and forwarding it to appropriate port transparently, just care about the next hop, so it is a layer 2 action.
Examples of Layer 2 devices: NIC, Bridge and Switch.
Hierarchical network design with access, distribution and core layers.
The benefits of the hierarchical network design.
Understanding hierarchical network design principles such as network diameter, bandwidth aggregation and redundancy.
The definition converged network.
Understanding different switch features:
1. Form Factors (Fixed, Modular and Stackable)
2. Performance (Port Density)
3. Performance (Forwarding Rates & Link Aggregation)
4. Power over Ethernet
5. L3 Functions
Access Layer Switch Features.
Distribution Layer Switch Features.
Core Layer Switch Features
Features of Cisco Catalyst Switches
Presentation from October 2012 RTI Technical Road Show.
Agenda Highlights:
How the DDS standard fosters information sharing and interoperability across systems of systems while driving down development, integration, maintenance, upgrade and acquisition costs
The latest 5.0 release of RTI's DDS solution and future roadmap, including enhanced security, support for integration patterns common in C2 systems, FAA DO-178C Level A certification, and DDS standardization initiatives
RTI's new Open Community Source license, which provides free-of-charge access to RTI DDS and allows it to be freely shared across projects and organizations
Layer 3 provides services to exchange the individual pieces of data over the network between identified end devices.
Understanding L3 addressing, encapsulation, routing and decapsulation.
Routed protocols, routing protocols and other supporting protocols such as ICMP and ARP.
IP v4 Characteristics:
1. Connectionless
2. Media independent
3. Best effort
IP v4 Header:
1. Packet length
2. Type of Service
3. Time to Live
4. Protocol
5. Header checksum
6. Source address
7. Destination address
8. Fragment offset.
9. IHL
10. Type of Service
This document contains study of Peer to Peer Distributed system.Three Models of Distributed system.Such as Centralizes,Decentralized,Hybird Model and Pros and cons of these models. Skpye and Bit torrent architecture is also discussed.This tutorial can be very help full for those who are beginners.
IEEE - Standards.
Ethernet Logical Link Control (LLC) and Media Access Control (MAC) sublayers.
Historic Ethernet: The foundation for Ethernet technology in 1970 was Alohanet.
Early Ethernet Media used coaxial cable to connect computers in a bus topology. It was known as Thicknet (10BASE5) and Thinnet (10BASE2).
Ethernet Collision Management.
There are two styles of Ethernet framing: IEEE 802.3 (original) and the revised IEEE 802.3 (Ethernet).
CSMA/CD, Carrier Sensle, Multiple Access, Collision Detection, JAM Signal and Random Backoff.
Ethernet Timing and latency.
Bit Time: The period of time required for a bit to be placed and sensed on the media.
Slot Time: is the time it takes for an electronic pulse to travel the length of the maximum theoretical distance between two nodes.
Inter-frame spacing.
Ethernet Types
"OpenHPC is a collaborative, community effort that initiated from a desire to aggregate a number of common ingredients required to deploy and manage High Performance Computing (HPC) Linux clusters including provisioning tools, resource management, I/O clients, development tools, and a variety of scientific libraries. Packages provided by OpenHPC have been pre-built with HPC integration in mind with a goal to provide re-usable building blocks for the HPC community. Over time, the community also plans to identify and develop abstraction interfaces between key components to further enhance modularity and interchangeability. The community includes representation from a variety of sources including software vendors, equipment manufacturers, research institutions, supercomputing sites, and others."
Watch the video: http://wp.me/p3RLHQ-gKz
Learn more: http://openhpc.community/
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
Trends in Systems and How to Get Efficient Performanceinside-BigData.com
In this video from Switzerland HPC Conference, Martin Hilgeman from Dell presents: HPC Workload Efficiency and the Challenges for System Builders.
"With all the advances in massively parallel and multi-core computing with CPUs and accelerators it is often overlooked whether the computational work is being done in an efficient manner. This efficiency is largely being determined at the application level and therefore puts the responsibility of sustaining a certain performance trajectory into the hands of the user. It is observed that the adoption rate of new hardware capabilities is decreasing and lead to a feeling of diminishing returns. This presentation shows the well-known laws of parallel performance from the perspective of a system builder. It also covers through the use of real case studies, examples of how to program for energy efficient parallel application performance."
Watch the video: http://wp.me/p3RLHQ-gIS
Learn more: http://dell.com
and
http://www.hpcadvisorycouncil.com/events/2017/swiss-workshop/agenda.php
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
Switching: means receiving data on a certain port and forwarding it to appropriate port transparently, just care about the next hop, so it is a layer 2 action.
Examples of Layer 2 devices: NIC, Bridge and Switch.
Hierarchical network design with access, distribution and core layers.
The benefits of the hierarchical network design.
Understanding hierarchical network design principles such as network diameter, bandwidth aggregation and redundancy.
The definition converged network.
Understanding different switch features:
1. Form Factors (Fixed, Modular and Stackable)
2. Performance (Port Density)
3. Performance (Forwarding Rates & Link Aggregation)
4. Power over Ethernet
5. L3 Functions
Access Layer Switch Features.
Distribution Layer Switch Features.
Core Layer Switch Features
Features of Cisco Catalyst Switches
1. Robert Franklin
3055 West Clyde Place
Denver, CO 80211
303-748-9114
wobfranklin@gmail.com
Profile
20 years of experience in all things UNIX and networking
Experience
June 2016 - Present
Shutterstock, Denver - Senior Network Engineer
● Senior Network Engineer for all of the Shutterstock network, including design, procurement,
installation, documentation, capacity planning, and operation of three co-location facilities
across the US (Boston MA, Secaucus NJ, Dallas TX). Network gear deployed Juniper MX
for core routing, Juniper SRX for firewall/VPN, Juniper EX for core and distribution
switching, Brocade and A10 networks for load balancing. Protocols used: OSPF, BGP,
LACP, VRRP, IPSEC VPNs, and general vlan switching/routing.
● Lead Engineer for Shutterstock’s Next Generation datacenter network. This includes design
and documentation of a Layer3 IP fabric that supports Layer2 overlay networks (VXLAN,
VMware NSX)
● Lead Engineer for Shutterstock’s On Premises DDoS mitigation service.
● Lead Engineer for Application Aware Network Performance Monitoring services using
Riverbed, gigamon, and Ixia taps.
● Senior Engineer for all layer7 load balancing of Shutterstock’s network. Vendors used: A10
networks, Brocade ADX
January 2011 - December 2015
Catalyst Systems Repository, Denver - Senior Network/Systems Engineer
● Lead Network Engineer for all of the Catalyst network, including design, procurement,
installation, documentation, capacity planning, and operation of five co-location facilities in
two countries (U.S., Japan). Network gear deployed - Juniper SRX series for core routing and
VPN services, Juniper Netscreens for firewalling, and Brocade FCX and ICX series for core
switching. Protocols used: OSPF, BGP, Spanning Tree, VRRP, IPSEC VPNs, LACP, and general
vlan switching/routing.
● Lead Engineer for all network support services, including RADIUS, DHCP, performance
monitoring (cacti),
● UNIX (Linux RedHat) DNS (BIND) engineer for the authoritative DNS servers for
catalystsecure.com
2. 2
● Lead BGP peering engineer for the Catalyst customer facing network, including configuration,
documentation, monitoring, capacity planning, and coordination with the upstream ISP
providers.
● Lead Engineer for all layer 7 load balancing services. This included design, procurement,
testing, implementation and documentation of multiple layer7 load balancing network devices
spread across five datacenters. Vendors included Brocade ADX and A10 Networks. Load
balanced applications included web services, FTP, and internal API middleware applications.
● Lead deployment engineer for datacenter expansions. This included design, procurement,
implementation and documentation of the datacenter networks. Duties included rack layout,
cabling, installation, configuration, and operation of the core network, customer facing
network, and all server connectivity
● Lead VMWare network engineer. Duties included configuration of vmware guests and
vswitch configuration for multiple HP C7000 blade chassis servers spread across multiple
datacenter locations. Also included design, configuration, installation, and deployment of
virtualized (A10 networks) load balancers in a VMWare host configuration.
● Lead performance troubleshooter. Included both network and UNIX system bottleneck
identification using standard UNIX system tools (vmstat, systat, iostat, SNMP monitoring,
SAR) and network performance investigations using tcpdump and wireshark.
● Deployed scripts for automation and configuration management for Linux systems, including
start-up scripts, log file sifting, and other text file manipulation using bash shell scripting tools.
February 2004 - December 2010
WildBlue Communications (startup)/ ViaSat, Greenwood Village - Senior Staff
Engineer, Network Architecture
● Lead UNIX (Sun Solaris) Engineer for all ISP services, including DHCP, DNS, TFTP, FTP. Duties
included capacity planning, design, documentation, systems tuning and management of all
deployed UNIX systems at the satellite gateway head-ends.
● Lead IP Security/Firewall (Juniper Netscreen)/VPN engineer for the DOCSIS satellite based
gateways. This included security specifications, design, vendor selection, procurement,
installation, capacity planning, and operations support for the gateway in-band and
out-of-band management networks.
● Lead UNIX (Sun Solaris) Engineer for all satellite gateway DNS services, including customer
facing recursive services for a DOCSIS satellite based broadband residential ISP. Duties
included lab proof of concept, performance analysis, software and hardware selection,
capacity planning, deployment, layer7 load balancing, documentation, and troubleshooting of
DNS services.
● Lead IP Engineer for all network load-balancing and web-caching architecture. Duties
included lab proof of concept work, performance analysis of web based protocols over high
3. 3
latency satellite, vendor selection, procurement of caching technology (BlueCoat Systems),
documentation along with care and feeding of the caching system. This also included TCP
stack performance tuning at both the satellite gateway head end servers and client web
browser performance tuning for optimal high latency (500ms RTT minimum) performance.
● Lead engineer for web performance. Duties included extensive lab testing in conjunction with
The Measurement Factory’s (http://www.measurement-factory.com, lead developers of
squid) web polygraph and ongoing performance monitoring of the WildBlue residential
network
● Lead layer7 load balancing (Foundry/Brocade) network engineer for the DOCSIS satellite
gateway headends. Duties included lab proof of concept work, performance analysis, vendor
selection, procurement and deployment of load balancers. Load balanced applications
included the WildBlue web caching system, and DNS services for both the web caching system
and residential end users.
● Design and Installation Team Member for over ten DOCSIS-based satellite-to-internet
gateways across North America. Responsibilities included design, lab testing, vendor
selection, installation, configuration, implementation, capacity planning, remote monitoring,
and documentation of the gateway networks and all ISP-related support services. Grew the
system from lab only development system from scratch to a production system with over
450,000 residential customers.
● Lead Systems Engineer for the DOCSIS satellite, web performance and bandwidth
measurement system. This included over 250 automated, embedded UNIX systems connected
to S-DOCSIS modems spread across multiple DOCSIS carriers/beams across North America
(US and Canada). Responsibilities included design, documentation, lab testing, procurement,
system automation, and monitoring of the system. This included bash and perl scripting to
automate configurations, text file parsing, and overall system management via cfengine
● Member of engineering residential broadband performance investigation and troubleshooting
team. Duties included researching sub-par performance on the DOCSIS carriers/satellite
beams and finding solutions using data gathered from the performance monitoring system,
packet captures and analysis using tcpdump and wireshark.
● Network architect team member. Satellite gateway networks based on Foundry/Brocade
BigIron8000 for core switching, and Juniper M series for core routing, including OSPF, BGP,
VRRP, and standard vlan switching/routing.
October 2002 - January 2004
ViaWest, Denver - Senior Network Engineer
● Lead Engineer for third shift. Duties included maintenance window documentation,
configuration, and deployment of new configurations, systems, and customer services for a
regional (SouthWest) datacenter and traditional ISP. Duties also included capacity planning
and troubleshooting of various ISP services including residential DSL, business T1s and DS3s,
frame relay circuits, backbone routing architecture, and multi-peer BGP design and
implementation.
4. 4
● Lead Engineer on datacenter integration. Duties included network integration design,
maintenance window configuration and deployment to tie in newly acquired datacenter
networks and systems.
● Lead Engineer for network monitoring. This included UNIX (Sun Solaris) system performance
tuning, management, and deployment of nagios and MRTG.
● Senior UNIX (Sun Solaris) engineer for all ISP based services, including customer facing DNS
services and authoritative DNS services using BIND, along with RADIUS and TACACS+
administration.
May 1999 - April 2002
High Speed Access (startup), Littleton - Senior Systems Engineer
● Lead UNIX (Sun Solaris)Engineer for all DOCSIS-based ISP services. Duties included lab
testing, configuration, capacity planning, and automation of Cisco CNR, the DHCP and TFTP
cable modem and CPE provisioning system.
● Lead Engineer for customer email and web hosting platform, including network and systems.
Duties included lab testing, capacity planning, vendor selection, performance tuning,
installation, and documentation. Email system was based on a Communigate Pro cluster on
Sun/Solaris servers with a Network Appliance storage system. Front end networking was
based on Foundry ServerIron layer7 switching/load balancing chassis systems. Web servers
were Sun/Solaris based running apache.
● Lead UNIX (Sun Solaris) engineer for both recursive customer facing DNS (BIND, djbdns)
services and authoritative DNS services. Lead DNS engineer for DNS services for the web and
email farms.
May 1998 - May 1999
eGlobe (startup), Denver - Senior Systems Engineer / Network Engineer
● Network Engineer for frame relay and dial-up network in over 80 countries. This included
procurement, configuration, provisioning, documentation and operation of the network.
● Lead UNIX (Sun Solaris) Engineer for customer email and web hosting platforms. Duties
included administration of both sendmail and apache on Sun Solaris hardware.
● Lead UNIX (Sun Solaris) engineer for both recursive customer facing DNS services and
authoritative DNS services
March 1997 - April 1998
USA.NET (startup), Colorado Springs - Senior Systems Administrator
● Lead UNIX Administrator for free email system with over 5 million users. Duties included
installation and administration of a 100+ Sun Solaris server platform, installation,
configuration and administration of qmail and apache web services.
● Lead storage administrator for the Oracle customer database and email storage services.
Duties included installation and configuration of Veritas Volume Manager and related
filesystem on EMC Symmetrix storage arrays. This also included Solaris performance tuning
for optimum I/O to the storage arrays via NFS and Oracle’s database files.
5. 5
● Datacenter Network Engineer. Duties included installation and configuration of the core
network based on a Cisco 6500 chassis, with cisco 4000M series routers peering with
upstream providers.
● UNIX (Solaris) administrator for the email platform’s DNS services
July 1995 - March 1997
Belen Consolidated Schools, Belen, NM - Senior Systems Administrator
● Lead UNIX administrator for a k-12 school system. Duties included installation and operation
of the web and email server running sendmail, apache, and BIND for DNS services.
● Lead network engineer for one of the first spread spectrum wireless deployments in the US.
This included IP network routing between 9 separate school buildings over point-to-point
wireless links. Duties also included installation and configuration of a cisco 2514 router IP
connectivity over a 56k frame-relay circuit to a regional ISP
Education
August 1994 - May 1995
University of New Mexico, Albuquerque, NM - Comp Sci
Technical Skills
● Senior-level systems engineer of *BSD, Linux, Solaris systems including automation,
administration and performance tuning
● Senior level DNS engineer for residential and regional ISP services including BIND and
unbound, nominum (lab testing and architecture only)
● Network design and architecture of carrier class ISP networks and ASP datacenter networks
● Documentation - Standard Operating Procedures creation, Best Practices and HOW-TOs,
wikis, diagrams
● System and network security, including OS, VPN, and firewalls
● Monitoring OS and network devices, SNMP, Nagios, Cacti
● Source Code Management - CVS, subversion, git
● Automated system installation and management - jumpstart, custom install scripts, cfengine,
puppet
● ISP services - DNS (BIND, unbound, djbdns), DHCP (isc-dhcpd, Incognito, Cisco CNR), FTP
(pure-ftpd, vsftpd, ftpd), TFTP (tftpd, Incognito, Cisco CNR) HTTP/S (apache, tiny httpd),
SMTP (sendmail, qmail, postfix), Web Caching (BlueCoat, squid), RADIUS
● Network devices - Cisco, Juniper/Netscreen, Brocade, A10, HP, Dell, BlueCoat, Console
servers, Motorola BSR 64000
● Systems - Dell, HP, Sun/Oracle, custom embedded/hardened systems
● Virtualization technologies - vmware (ESX, ESXi, vSwitch networking), Xen, VirtualBox,
BSD/Solaris Jails
● Network protocols - Ethernet, ATM, Packet over SONET, DSL, DOCSIS, S-DOCSIS, VLANs,
LAGs, STP, PVSTP, ARP, RARP, ICMP, OSPF, ISIS, BGP, HSRP, VRRP, IP, TCP, UDP, XTP,
vmware vSwitch, VirtualBox