Transcript QoS in an Ethernet World

QoS in an Ethernet world
Bill Lynch
Founder & CTO
www.procket.com
QoS
• Why is it needed? (Or is it?)
• What does it do? (Or not do?)
• Gotchas….
• Why is it hard to deploy?
2
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Triple play data networks
VOD, CONF, Data services
Interface content
mirroring for
security
requirements
High-speed Ethernet Edge
• Assured QoS
• DOS prevention
Edge
PE
Distribution
IP or MPLS or
λ Core
CE
CE
VPN A
VPN B
CE
PE
VPN A
Broadband Home
Centralized
Headend
•Video, voice, data over ethernet.
•QoS across thousands of subscribers
•SLAs and differential pricing
Headend
Computational
Particle Physicist
Headend
3
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Triple play data characteristics
•Voice
• Many connections
• Low BW/connection
• Latency/jitter requirements
•Video
• Few sources
• Higher BW
• Latency
•Data
• Many connection
• Unpredictable BW
• BE generally okay
•Computational particle physicist
• Very high peak BW & duration
• Very few connections
4
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Router QoS
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
5
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Router QoS
QoS == which packet goes first
Only matters under congestion
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
6
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Router QoS
Inherent packet jitter
Worse: N simultaneous arrivals
Bad: Per hop!
Worse: Bigger MTU
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
7
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Inherent jitter (per hop!)
1500B latency (us)
Fundamental conclusion:
QoS more important at edge
1000.0
Edge also more likely
to congest
120.0
100.0
FE
OC-12
GE
19.2
12.0
10.0
OC-12
4.8
OC-192
1.2
1.0
0.1
1
10
8
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Gotchas….
• Already no guarantees from simultaneous arrival…
… but hope the total worst case is < 10ms?
• And what if your router wasn’t perfect?
9
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
What is Queue Sharing?
Queue Sharing is when multiple physical or switch fabric
connections must share queues.
Example: Each input linecard has two queues for each output linecard.
All packets in a shared queue are treated equally.
Physical Port
Physical Port
Physical Port
Physical Port
HI Queue
HI Queue
LO Queue
LO Queue
Physical Port
Physical Port
Physical Port
Physical Port
10
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
What is Head of Line Blocking?
When an output linecard becomes congested, traffic
becomes congested on the input linecard
Traffic control (W/RED) must be performed at input VOQ.
Physical Port
HI Queue
HI Queue
LO Queue
LO Queue
Physical Port
Physical Port
Physical Port
11
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
What is Head of Line Blocking?
The output linecard cannot process all of the output traffic.
Because all traffic in a shared queue (VOQ) is treated equally,
we have affected traffic on the uncongested port.
Physical Port
HI Queue
HI Queue
LO Queue
LO Queue
Physical Port
Physical Port
Physical Port
12
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Queue Sharing Test Results
Congested port (Flows C, D, E) remained at 100% throughput
Uncongested (Flows A, B) were penalized because of Queue Sharing
13
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
The effects of Queue Sharing
With the presence of Queue Sharing, congestion can
severely affect the performance of non-congested ports
Congestion is caused by:
Topology Changes
Routing Instability
Denial of Service Attacks
High Service Demand
Misconfiguration of systems or devices
14
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Architectures - PRO/8000
Only one queuing location exists in the entire system
36,000 unique hardware queues
Protected bandwidth on a queue
Incoming packets are immediately placed into a unique
output queue
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Centralized Shared Memory
Switch Fabric
Physical Port
15
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Architectures - PRO/8000
Only one queuing location exists in the entire system
Over 36,000 unique hardware queues
Bandwidth is protected on a per-queue basis
Incoming packets are immediately placed into a unique
output queue
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Centralized Shared Memory
Switch Fabric
Physical Port
16
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Architectures - PRO/8000
Traffic control (W/RED) is performed on each output queue
individually
Protected bandwidth for every single queue
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Centralized Shared Memory
Switch Fabric
Physical Port
17
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Pro/8812 Test Results
Congested port (Flows C, D, E) remained at 100% throughput
Uncongested (Flows A, B) remained at 100% throughput
18
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Triple play data characteristics
•Voice
• Many connections
• Low BW/connection
• Latency/jitter requirements
•Video
• Few sources
• Higher BW
• Latency
•Data
• Many connection
• Unpredictable BW
• BE generally okay
•Computational particle physicist
• Very high peak BW & duration
• Very few connections
19
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Network Qos architectures
Network
Predictability
QoS
PSTN
50 years
fixed BW
TDM
Cable MSO
50 years
transmit only
Provision and
broadcast
Data
Evolving
Over-provision
20
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
QoS Deployment Issues
•Political
• Peers
•Equipment
•
•
•
•
•
QoS is end to end
Many queues/port
Many shapers/port
Fast diffserv/remarking
Computation expense
•Operational
• Must deploy everywhere
• Must police at the edge
•Commercial
• Easier short term solutions to
problems
• Cheaper alternatives
•Applications
• Not tuned or aware
• QoS not ‘required’ for the
application
•Geographical
• Last mile technologies
• Single provider network
• Green field deployments
21
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Summary
• Triple play requires QoS
• Services drive quality
• Most routers aren’t perfect
• Shared queues mean you can’t provision a port independently
• Political and deployment problems remain
• Some geographic areas better suited
22
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
23
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Never underestimate the power of Moore’s Law
SC
LCU
297 sq mm
(17.26mm x 17.26mm)
30.5M transistors
47M contacts
50KBytes of memory
425 sq mm
(20.17mm x 21.07mm)
137M transistors
188M contacts
950KBytes of memory
Striper
429 sq mm
(20.17mm x 21.29mm)
156M transistors
265M contacts
1.2MBytes of memory
NPU
Architecture
429 sq mm
(20.17mm x 21.29mm)
214M transistors
400M contacts
2.6MBytes of memory
MCU
GA
389 sq mm
(19.05mm x 20.4mm)
106M transistors
188M contacts
1.2MBytes of memory
225 sq mm
(15.02mm x 15.02mm)
83M transistors
136M contacts
900KBytes of memory
24
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
NPU – 40G QoS lookups
VLIW systolic Array
• Packet advances
every cycle
• Named bypassing
• > 200 processors
• 4 ops/cycle/processor
PxU
LxU
FTSRAM
• 12 loads every cycle
IPA
PBU
pacman
• (1Tb memory BW)
• 36 loads/packet
QxU
25
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
NPU
VLIW systolic Array
• Normal instruction
set
PxU
• Arithmetic
• Logical
• Branch
LxU
FTSRAM
• Load
• Simple
programming model
• Deterministic
performance
IPA
PBU
pacman
QxU
26
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Memory Controller – Service Level Queueing
• High BW
• 16 DRAM chips
• independent memory banks
• BW dist. across banks
• 36K queues
• Memory management
• Write-once multicast
• Preserve ordering
27
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Basic Router Architecture Elements
Linecard
Switch Fabric
Linecard
Three Classes of Switch Fabric Architecture
- Input Queued (IQ)
- Output Queued (OQ)
- Combined Input/Output Queued (CIOQ)
28
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Input Queued (IQ) Fabrics
Input Linecard
Switch Fabric
Ouput Linecard
Input Queued Switch Fabrics:
Inefficient use of memory
Require Complex Scheduling
29
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Combined Input/Output Queued (CIOQ) Fabrics
Input Linecard
Switch Fabric
Ouput Linecard
CIOQ Switch Fabrics:
Generally with point-to-point fabric in the middle
(Crossbar, multi-stage (clos), torus)
Requires Complex Scheduling
Queues shared to reduce complexity
30
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Fabrics
Input Linecard
Switch Fabric
Ouput Linecard
OQ Switch Fabrics:
Require extremely high speed memory access
Do not share queues
Efficient multicast replication
Protected bandwidth per queue
31
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Terabit Centralized Shared Memory
Routers
April 20, 2004
Bill Lynch
CTO
www.procket.com
Whither QoS?
April 20, 2004
Bill Lynch
CTO
www.procket.com
Concurrent Services
VOD, CONF, Data services
Interface content
mirroring for
security
requirements
High-speed Ethernet Edge
• Assured QoS
• DOS prevention
Edge
PE
IP
MPLS
λ
Distribution
CE
CE
VPN A
VPN B
CE
PE
VPN A
Broadband Home
Centralized
Headend
•Video, voice, data over ethernet.
•QoS across thousands of subscribers
•SLAs and differential pricing
Headend
Research,
Education, Grid,
Supercomputing
Headend
34
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
35
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
(More Bill’s Slides Here)
• (As much detail on the switch fabric and chips as you are comfortable saying in
a multi-vendor environment!)
• No scheduling
• 36K service level queues
• NPU for fast lookup, policing, shaping
• SW abstraction based on service performed, not provided knobs
• Many, many, many DRAM banks. However, ½ as many as CIOQ architectures.
• 40G NPU for line rate
• Policing
• Remarking
• DA, AS, other lookup
• SW interface focus on service, not knobs.
36
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
(Insert Bill’s Slides Here)
• Self Introduction
• Problem Statement (Bill)
• "Layer 3 QoS at the right scale price is elusive"Throwing more bandwidth at lower layers
only makes networking researchers commodity bandwidth brokers. Also that is fine for
R&E but commercially that is too expensive, so there appears to be a growing disconnect
between R&E and commercial.It will be important not to slam the current L2/L1 vogue lest
we upset the locals :)
• Numerous commercial implementations starting now
• Single network country
• High BW to home
• Triple play
• Assertion (Bill)
• "System Architecture greatly contributes to the proper operation of network wide
QoS"Current system architecture are completely unfocused on network wide QoS, and
focused on per-hop-behaviors. This forces networkers to tweak 100 knobs to get the
desired behavior. Why not architect the system to protect a flow through the router, so
that behaviors are predictable in every circumstance?
• End 2 end. Any problem exacerbated by TCP.
CONFIDENTIAL
37
© 2004 Procket Networks, Inc. All rights reserved.
Abilene Network Map
Source: http://abilene.internet2.edu/new/upgrade.html
38
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Internet Growth Predictions
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
“117% YEARLY GROWTH THROUGH 2006”
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
“VIDEO WILL DRIVE TRAFFIC GROWTH
OVER THE NEXT 10 YEARS”
Source: Yankee Group April 2004
39
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Network Reference Design
Single Element Core (Cluster)
Interdomain QoS
Peers
Concurrent Services Edge
Intradomain QoS
40
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
PRO/8000TM Concurrent Services Routers
Highest performance and density
960Gbps
2 per rack
Ultra-compact
80Gbps
8 per rack
41
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
PRO/8000 Series Logical Architecture
#
Procket VLSI
Forwarding Plane
CP
Route Processors
(1+1)
CP Control Plane
1
1
1
1
5
CP
2
4
3
Line Card
Switch Cards (2+1)
Media
Adapters
Line Card
5
5
5
Media
Adapters
• Fully redundant Switch Cards
and Route Processors
• All components hot-swappable
in-service
• No single point of failure
• Strictly non-blocking
42
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Basic Router Architecture Elements
Linecard
Switch Fabric
Linecard
Three Classes of Switch Fabric Architecture
- Input Queued (IQ)
- Output Queued (OQ)
- Combined Input/Output Queued (CIOQ)
43
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Input Queued (IQ) Fabrics
Input Linecard
Switch Fabric
Ouput Linecard
Input Queued Switch Fabrics:
Inefficient use of memory
Require Complex Scheduling
44
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Combined Input/Output Queued (CIOQ) Fabrics
Input Linecard
Switch Fabric
Ouput Linecard
CIOQ Switch Fabrics:
Generally with point-to-point fabric in the middle
(Crossbar, multi-stage (clos), torus)
Requires Complex Scheduling
Queues shared to reduce complexity
45
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Fabrics
Input Linecard
Switch Fabric
Ouput Linecard
OQ Switch Fabrics:
Require extremely high speed memory access
Do not share queues
Efficient multicast replication
Protected bandwidth per queue
46
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
What is Queue Sharing?
Queue Sharing is when multiple physical or switch fabric
connections must share queues.
Example: Each input linecard has two queues for each output linecard.
All packets in a shared queue are treated equally.
Physical Port
Physical Port
Physical Port
Physical Port
HI Queue
HI Queue
LO Queue
LO Queue
Physical Port
Physical Port
Physical Port
Physical Port
47
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
What is Head of Line Blocking?
When an output linecard becomes congested, traffic
becomes congested on the input linecard
Traffic control (W/RED) must be performed at input VOQ.
Physical Port
HI Queue
HI Queue
LO Queue
LO Queue
Physical Port
Physical Port
Physical Port
48
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
What is Head of Line Blocking?
The output linecard cannot process all of the output traffic.
Because all traffic in a shared queue (VOQ) is treated equally,
we have affected traffic on the uncongested port.
Physical Port
HI Queue
HI Queue
LO Queue
LO Queue
Physical Port
Physical Port
Physical Port
49
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Queue Sharing Test Results
Congested port (Flows C, D, E) remained at 100% throughput
Uncongested (Flows A, B) were penalized because of Queue Sharing
Traffic on adjacent ports was dropped!
50
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Architectures - PRO/8000
Only one queuing location exists in the entire system
Over 36,000 unique hardware queues
Protected bandwidth down to DS3 granularity
Incoming packets are immediately placed into a unique
output queue
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Centralized Shared Memory
Switch Fabric
Physical Port
51
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Architectures - PRO/8000
Only one queuing location exists in the entire system
Over 36,000 unique hardware queues
Protected bandwidth down to DS3 granularity
Incoming packets are immediately placed into a unique
output queue
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Centralized Shared Memory
Switch Fabric
Physical Port
52
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Output Queued Architectures - PRO/8000
Traffic control (W/RED) can be performed on each output
queue individually!
Protected bandwidth for every single queue
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Physical Port
Centralized Shared Memory
Switch Fabric
Physical Port
53
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Multicast Scaling and Performance by Design
Content
Incoming Line Card
Media Adapters
Centralized Shared
Memory Switch Fabric
1. One copy of packet
written into memory
Outgoing
Line Cards
2. Output Line Cards read copy of
packet out of memory
3. Copy packet to each outgoing
interface
54
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
State-of-the-art Networking Software
Embedded self-diagnostics
Lightweight kernel
In service upgrades
Fully modular
Automatic image rollback
Memory protection
Simple to extend
Inherent fault isolation
Modular forwarding code
Restartable processes
Rapid recovery from failures Built in portability
Fast messaging between processes
55
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Portable, Lightweight Kernel
Portability Ensures Longevity & Consistency of System Software,
Lightweight Operating System Maximizes System Stability
Lightweight Kernel to handle
scheduling and memory allocation
• Portability of PRO/1 MSE is built in
• System software features can easily be moved to new platforms
• Stripped down to essential functions to maximize
stability
• No networking functions or services can crash the system
• Designed for mission critical applications
56
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Modular versus Monolithic Alternatives
Fully modular
BGP
OSPF
…
IGMP
PIM
…
SNMP
CLI
Intelligent
…Service
Agent
Other
System Manager
Lightweight Kernel
Semi-monolithic
CLI Interfaces
Monolithic
BGP
IS-IS, OSPF
PIM, MSDP, SSM, IGMP
Kernel
CLI
Interface Mgr
BGP
IS-IS, OSPF
PIM, MSDP, SSM, IGMP
Kernel
57
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Improve Network Availability, Simplify Operations
In-service software upgrades increase network availability
and simplify network operations
New SNMP
Package
Base
Release
BGP
OSPF
…
IGMP
PIM
…
Intelligent
SNMP
SNMP
CLI
…Service
Other
Agent
1.
2.
3.
Package
installed
and running
Procket Package Manager checks compatibility of base release and SNMP package
While SNMP package is installed, all protocols continue to operate
Once installed, SNMP can be restarted using the new package
BGP
OSPF
…
IGMP
PIM
…
New
SNMP
Package
…
CLI
Intelligent
Service
Agent
Other
58
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Software Architecture
• Each protocol runs as a separate process
• Uses multiple POSIX threads for scheduling tasks
• Uses private memory for local data structures
• Uses well documented APIs to service other processes
• Uses shared-memory when offering read-only API
service to other processes
• Run-to-completion thread scheduling
• Table managers run as separate processes
59
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
IPC Example
URIB
OSPF
- URIB writes
- Learns Route
- Adds to URIB
to memory
- Uses mq IPC
R/W
R/W
IP
- Packet arrives
- Route lookup
- Uses direct
read
R/W
URIB
API
URIB
API
OSPF
URIB
IP
routing
table
Shared-Memory
60
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Programmable VLSI Forwarding Engine
• Facilitates line-rate forwarding of IPv4, IPv6, and MPLS
traffic
• New services with software downloads rather than
hardware upgrades
• Support for IPv6
• TTL checking in hardware (for ACLs, GTSM …)
• Multiple priority queues for RP destined traffic
• capable of modifying queue priority for various types of control
traffic
• Multicast (PIM-SM) support does not need special
media-adaptors
61
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
CLI Highlights
• Familiar ‘look and feel’ reduces OpEx
• Operations mode examples
show ip bgp summary
show ip interface brief
show ip ospf neighbors
show isis database
show ip mroute
• Configuration mode examples
router bgp 100
log-neighbor-changes
neighbor 10.1.1.1 remote-as 200
dont-capability-negotiate
address-family ipv4 unicast
policy remove-martians in
• Support for deferred configuration “commits”
62
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Other salient software features
• Powerful policy specification framework
• intuitive syntax - similar to structured programming languages
• support for “chaining” actions
• Service oriented, modular QoS configuration
• Dynamic RP queue prioritization for known BGP peers
• Ability to constrain debug output using “debug-filters”
• Conservative defaults
• unnecessary services are disabled (only ‘ssh’ is on)
• Digitally signed software packages for verifying source and integrity of
contents
• Intelligent service-agent for pro-active health monitoring
63
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Procket PRO/Silicon Technology
Highest performance and density
960Gbps
2 per rack
Ultra-compact
80Gbps
8 per rack
64
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Procket PRO/Silicon Technology
65
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Procket PRO/Silicon Technology
• World’s fastest packet processors
• First 40 Gbps network processor (2002)
• Record bandwidth density
• 6-chip family
• Most flexible platform
• Unmatched programmability
enables new services
• Long lifetime
• Enhanced reliability
• Highest level of silicon integration
NPU
214 Million Transistors
66
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Never underestimate the power of Moore’s Law
SC
LCU
297 sq mm
(17.26mm x 17.26mm)
30.5M transistors
47M contacts
50KBytes of memory
425 sq mm
(20.17mm x 21.07mm)
137M transistors
188M contacts
950KBytes of memory
Striper
429 sq mm
(20.17mm x 21.29mm)
156M transistors
265M contacts
1.2MBytes of memory
NPU
Architecture
429 sq mm
(20.17mm x 21.29mm)
214M transistors
400M contacts
2.6MBytes of memory
MCU
GA
389 sq mm
(19.05mm x 20.4mm)
106M transistors
188M contacts
1.2MBytes of memory
225 sq mm
(15.02mm x 15.02mm)
83M transistors
136M contacts
900KBytes of memory
67
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
40 Gbps NPU
•
•
•
•
•
•
•
•
VLIW systolic array
375 MHz
125 Mpps
2856 min ops/packet
37 min loads/packet
255 meters
256K GPCID
Programmable features
• Parsing, Lookup, PCL
QOS, Accounting, IPv6
68
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
40 Gbps NPU
VLIW systolic Array
• Packet advances
every cycle
• Named bypassing
• > 200 processors
• 4 ops/cycle/processor
PxU
LxU
FTSRAM
• 12 loads every cycle
IPA
PBU
pacman
• (1Tb memory BW)
• 36 loads/packet
QxU
69
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
40 Gbps NPU
VLIW systolic Array
• Normal instruction
set
PxU
• Arithmetic
• Logical
• Branch
LxU
FTSRAM
• Load
• Simple
programming model
• Deterministic
performance
IPA
PBU
pacman
QxU
70
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Memory Controller – Service Level Queueing
• High bandwidth
• 16 DRAM chips
• independent banks
• BW across banks
• 36K queues
• Memory management
• Write-once multicast
• Preserve ordering
TQ
HQ
MLT
MQCC
QCC
COHB
SCIB
71
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
PRO/Silicon Technology
Advanced Silicon Development
Max Density
Max Speed
Max Reliability
Min Power
Min Cost
Procket facilities provide
complete control over chip design
Architecture
ASIC designer
Logic
Gates
Layout
Fab
Custom
Package
ASIC vendor
72
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
Procket PRO/Silicon Technology
Reliable
High integration
Space efficient
Highest Density
Architecture
OPEX savings
Highest power Efficiency
Future proof
Full programmability
73
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.
PRO/8000TM Concurrent Services Routers
Highest performance and density
960Gbps
2 per rack
Ultra-compact
80Gbps
8 per rack
74
CONFIDENTIAL
© 2004 Procket Networks, Inc. All rights reserved.