FireFly: A Reconfigurable Wireless Datacenter Fabric using - - PowerPoint PPT Presentation

▶

Aug 25, 2022 251 likes •572 views

FireFly: A Reconfigurable Wireless Datacenter Fabric using Free-Space Optics Navid Hamedazimi, Zafar Qazi, Himanshu Gupta, Vyas Sekar, Samir Das, Jon Longtin, Himanshu Shah, Ashish Tanwer ACM SIGCOMM 2014 Datacenter network design is hard!

SLIDE 1

FireFly: A Reconfigurable Wireless Datacenter Fabric using Free-Space Optics

Navid Hamedazimi, Zafar Qazi, Himanshu Gupta, Vyas Sekar, Samir Das, Jon Longtin, Himanshu Shah, Ashish Tanwer

ACM SIGCOMM 2014

SLIDE 2

Datacenter network design is hard!

Cost Performance Cabling Expandability Energy Cooling Adaptability

SLIDE 3

Existing Data Center Network Architectures

…

Over subscribed (e.g. simple tree) Augmented (e.g. cThrough)

…

Over provisioned (e.g. FatTree, Jellyfish)

…

SLIDE 4

Our Vision : FireFly

Coreless
Wireless
Steerable

ToR switch FireFly Controller

Steerable Links

SLIDE 5

Potential Benefits of This Vision

Cost Performance Cabling Expandability Energy Cooling Adaptability

Wireless Coreless Steerable

SLIDE 6

Challenges in Realizing the Vision

FireFly Controller ToR switch

Steerable FSOs

Steerable wireless links
Network Design
Network Management

FireFly shows this vision is feasible

SLIDE 7

Outline

Motivation
Steerable Wireless Links
Network Design
Network Management
Evaluation

SLIDE 8

Why FSO instead of RF?

RF (e.g. 60GHZ) FSO (Free Space optical)

Wide beam  High interference Limited active links Limited Throughput Narrow beam  Zero interference No limit on active links High Throughput

SLIDE 9

Today’s FSO

Cost: $15K per FSO
Size: 3 ft³
Power: 30w
Non steerable
Current: bulky, power-hungry, and expensive
Required: small, low power and low expense

SLIDE 10

Why Size, Cost, Power Can be Reduced?

Traditional use : outdoor, long haul

‒ High power

‒ Weatherproof

Data centers: indoor, short haul
Feasible roadmap via commodity fiber optics

‒ E.g. Small form transceivers (Optical SFP)

SLIDE 11

FSO Design Overview

SFP fiber optic cables

SLIDE 12

FSO Design Overview

SFP Diverging beam

SLIDE 13

FSO Design Overview

SFP Lens focal distance

large cores (> 125 microns) are more robust

Large core fiber optic cables Parallel beam lens Focusing lens Collimating lens

SLIDE 14

Steerability

Cost Size Power

Not Steerable

FSO design using SFP

Via Switchable mirrors

r Galvo mirrors

Shortcomings of current FSOs

SLIDE 15

Steerability via Switchable Mirror

A Ceiling mirror B C

Switchable Mirror: glass mirror
Electronic control, low latency

SM in “mirror” mode

SLIDE 16

Steerability via Galvo Mirror

A Ceiling mirror B C

Galvo Mirror: small rotating mirror
Very low latency

Galvo Mirror

SLIDE 17

FSO Prototype in Data center

Fiber holder and lens Mirror

SLIDE 18

FSO Link Performance

6 mm 6 mm

FSO link is as robust as a wired link

Effect of vibrations, etc.
6mm movement tolerance
Range up to 24m tested

SLIDE 19

Outline

Motivation
Steerable Wireless Links
Network Design
Network Management
Evaluation

SLIDE 20

How to design FireFly network?

Goals: Robustness to current and future traffic
Budget & Physical Constraints
Design parameters

– Number of FSOs? – Number of steering mirrors? – Initial mirrors’ configuration

Performance metric

– Dynamic bisection bandwidth

SLIDE 21

FireFly Network Design

# of FSOs = # of Servers
# of Switchable Mirrors = [10-15] for up to 512 racks
r
# of Galvo Mirrors = 1 per FSO
Mirror Configuration = Random graph
less than ½ the ports of FatTree

Projected Cost: 40% to 60% lower than FatTree

SLIDE 22

Outline

Motivation
Steerable Wireless Links
Network Design
Network Management
Evaluation

SLIDE 23

Network Management Challenges

Reconfiguration

– Traffic engineering – Topology control

Correctness during flux

ToR switch FireFly Controller Steerable FSOs Ceiling Mirror

SLIDE 24

FireFly Reconfiguration Algorithm

Joint optimization problem
Decouple

– Traffic engineering – Topology control

Above is done periodically
In addition: Trigger-based reconfiguration

– E.g. Create direct link for large flows

Massive ILP
Max-flow, greedy
Weighted Matching

SLIDE 25

Correctness Problems During Flux

Connectivity
Black Holes
Latency

A B A B A B C C C

SLIDE 26

Simple Rules To Ensure Correctness

Disallow deactivations that disconnect the network.
Stop using a link before deactivating it
Start using a link only after activating it
“Small” gap between reconfigurations

SLIDE 27

Outline

Motivation
Steerable Wireless Links
Network Design
Network Management
Evaluation

SLIDE 28

FireFly Evaluation

Packet-level
Flow-level (for large scale networks)
Evaluation of network in-flux
Evaluation of Our Heuristics

SLIDE 29

2 4 6 8 10

hotspot (8) hotspot (16) Uniform

fireFly cThrough Fattree

i

Throughput per server in Gbps Htsim simulator, 64 racks, three traffic patterns

FireFly is comparable to FatTree with less than ½ the ports Flow completion time better than FatTree

FireFly Throughput

SLIDE 30

Conclusions

Vision: Extreme DC network architecture

– Fully Steerable, No core switches, All-wireless inter-rack

Unprecedented benefits:

– No Cabling, Adapt to traffic patterns, Less clutter

Firefly shows a viable proof point

– Practical steerable FSO for datacenters – Practical network design and management heuristics – Close to fat tree performance over several workloads – Less than half of FatTree ports

Just a start .. Many directions for improvement