1. High Speed Optical Interconnect Project May08-06

2. Client: Lockheed Martin
Team Information
Client: Lockheed Martin
Team Members
Team Leader: Adam Jackson
Communication Coordinator:
Nick Ryan
Bader Al-Sabah
David Feely
Richard Jones
Faculty Advisor
Dr. Ahmed Kamal
Client Contacts
Aaron Cordes
Rick Stevens

3. Problem Statement Approach
At this time, the maximum real-world throughput of 10 Gbps network configurations is unknown.
Approach
Investigate a 10 Gbps network using fiber optical components
Test final system
Problem Statement: We need knowledge of the maximum actual throughput of a 10 Gbps network configuration.
Our Approach:
investigate implementing a 10 Gbps network using fiber optical components
Research various architectures to identify the best one for our purpose
The final design will then be tested to measure network characteristics.

4. Requirements Functional Non-Functional
The network shall be implemented with 10 Gbps components.
The team shall test the network to determine real throughput.
Non-Functional
Must be implemented before the class completes in May 2008.
Must determine financial feasibility of fiber optic components.
Our Requirements are to implement a network with 10 Gbps components, test the network to determine the real throughput, implement design before class completes, and determine the financial feasibility of the fiber optic components.

5. Functional Requirements Cont.
2.1 The team shall test the network to determine real-world bandwidth.
2.2 The team shall test the network to determine bandwidth efficiency.
2.3 The team shall test the network to determine switching time.
2.4 The team shall test the network to determine latency.
2.5 The team shall test the network to determine quality of service.
We will be testing to determine
real- world bandwidth
bandwidth efficiency
switching time
latency
and quality of service
The details of these tests will be discussed later

6. Project Risks A team member could leave the project
Insufficient funding for project completion
Component failure
Project Risks
Team member leaving – internship, illness
Insufficient funding for completion
Component failure – ESD, delivery issues

7. Deliverables
The team shall present the chosen design solution to LM for approval before implementation.
The team shall provide weekly reports to LM.
The team shall present the test results to LM.
The team shall deliver a final report and all documentation to LM.
The team shall deliver the prototype to LM at the completion of this project.
All deliverables shall be presented to Rick Stevens on or before May 1, 2008.
Deliverables:
Design solution before implementation
Weekly Reports
Test Results
Final Report and Documentation
Prototype
On or before May 1

8. Design Strategy Block Diagram
This is our overall design Strategy, We started with Market and technology Research, then came design, implementation and testing during which could polish our design until it fits our requirements and we arrive at our final product.
Design Strategy Block Diagram

9. Market Research µTCA PCI-E Advantages Disadvantages Advantages
Modular design allows for expansion
262.5 Gbps maximum throughput for Advanced Mezzanine Cards (AMC)
Disadvantages
AMC Network Interface Cards at 10 Gbps are not readily available
Costly components
Advantages
Readily available optical 10 Gbps NICs
Variety of 10 Gbps XFP Switches
Relatively low cost components
Disadvantages
Lack of PCI-E systems at ISU
Source:

10. Point-To-Point Configuration
Testing was completed with two systems directly connected
Used for testing bandwidth and bandwidth efficiency

11. Switched Configuration
Composed of three nodes and a Ethernet switch
Used for testing switching time, latency, and quality of service
Graphic inspired by previous HSOI team

12. Myricom 10G-PCIE-8A-R+E®
Hardware Setup
Myricom 10G-PCIE-8A-R+E®
10 Gigabit Ethernet PCIe Adapter with pluggable XFP optical interface (
Node 1
Node 2
Node 3
XFP Transceiver
XFP for each NIC
PCI-E x8 for NICs
Dell Optiplex 745 running Linux
Lab operating environment
PCI-E x8 or x16 bus
TigerSwitch 10G 8-Port Standalone XFP 10Gigabit Ethernet Managed Layer 2 Switch
SMC Networks, Inc.

13. Testing and Analysis Software
ping
Used as a latency measurement tool
iperf
Bandwidth measurement tool
tcpdump
Packet capture program
Linux Shell Scripts
Scripts will control and execute programs according to test specifications
Analysis Programs
C++ program used to analyze results
Need to mention PCI-E x8 NICs on previous slide

14. Test Specifications Bandwidth Measure maximum link usage
Bandwidth Efficiency
Calculate real data vs. transmission data
Latency
Measure end-to-end delay over a single link

15. Test Specifications Switching Time
Difference in latency in point-to-point vs. switched configurations
Quality of Service
Show the amount data received from each sending node for each endpoint node over time
Planned, not realized

16. Bandwidth Test Results
30 iperf runs
Command Line: iperf -p c
Max: 1.39 Gbps
Min: 1.38 Gbps
Avg: Gbps
Standard Deviation: Gbps
99% Confidence Interval: [ , ] Gbps

17. Bandwidth Efficiency Calculations
Header & CRC: 26 bytes
64b/66b Encoding

18. Latency Test Results 1000 ping runs ping -c 5 192.168.1.2
Max: ms
Min: ms
Avg: ms
Standard Deviation: ms
99% Confidence Interval: [ , ] ms

19. Resource Costs Resource Quantity Estimated Cost Actual Cost
Optical NICs 3
$3000 $
XFP Switch 1
$6500
N/A
XFP Transceiver 6
$3720
$2700
Fiber optic cables
$240
$130.94
Host System
$1500
Department Systems1
Total
$14960 $
1 ISU ECpE Department’s update of the Senior Design lab may cover this cost
2$10 Shipping charge for NICs and Transceivers added to total

20. Planned Work Breakdown
Personnel
Meetings
Research
Design
Implementation
Testing and
Debugging
Documentation
Website
Totals
Adam 80 20 8 32 10 1
171
Bader
Nick
David
Richard 15 30
178
Total
400
100 40 95
158 50 19
862

21. Actual Work Breakdown Summary of Individual Hours Personnel Meetings
Research
Design
Implementation
Testing and
Debugging
Documentation
Website
Totals
Adam
73.75 34 2 7
14.75
21.25
152.75
Bader 81 29 1
5.5
122
Nick
79.5
25.5 4
3.75
19.5
133.25
David
60.5 30
8.5 5 11
115
Richard
79.25 44
1.5 6 12
18.75
161.5
Total
374.05
162.5
7.5
17.5 35
69.25
684.5
Budgeted
400
100 40 95
158 50 19
862
Summary of Individual Hours

22. Project Completion Analysis
Estimated Percent Completion
Budgeted Hours
Budgeted Hours Completed
Meetings
100%
400
Research
95%
100 95
Design 40
Implementation
80.75
Testing
75%
158
118.5
Documentation 50
47.5
Website
97% 19
18.5
Used to calculate Budgeted cost of whatever nick is talking about
Table used to calculate Budgeted Cost of Work Performed

23. Project Analysis Budgeted Cost of Work Scheduled:
862 scheduled hours * $10 per hour = $8,620
Actual Cost of Work Performed:
actual hours * $10 per hour = $6,845.50
Budgeted Cost of Work Performed:
scheduled hours * $10 per hour = $8,002.50

24. Earned Value Analysis Cost Variance Cost Performance Index
$8, $6,735 = $1,267.5
Under budget
Cost Performance Index
$8,002.5 / $6,735 = 1.188
Schedule Variance
$8, $8,620 = $-617.5
Behind schedule
Schedule Performance Index
$8,002.5 / $8,620 = 0.928

25. Realized Risks Unable to obtain fiber optic switch
PCI-E slot not fully compliant
First system ran at x1
Second system was graphics only
Switch – ISU couldn’t fund, LM couldn’t fund, couldn’t rent/loan PCI-E – Tried 2 systems to get x8 speed

26. Accomplishments Researched 10 Gbps network architectures
Completed test scripts
Completed analysis scripts and programs
Executed copper tests and analyzed results
Executed limited fiber optic tests

27. Lessons Learned
How to coordinate buying components with various sources.
Prepare for unexpected sources of error.
Fully research the technologies before we use them.

28. Conclusions Creating an affordable fiber optic network is feasible.
Copper results illustrate proof of concept on testing approach.
Preliminary results using PCI-E x1 speeds show that bandwidth efficiency is poor unless parameters are changed.
Creating an affordable fiber optic network is feasible.
Consequent testing was inconclusive due to issues with PCI-E and budgetary constraints.
Copper results illustrate proof of concept on testing approach.
Preliminary results using PCI-E x1 speeds show that bandwidth efficiency is poor unless parameters are changed.
Latency, switching time, and QOS were all satisfactory.

29. Questions? ?