Download presentation
Presentation is loading. Please wait.
1
Router Construction II Outline Network Processors Adding Extensions Scheduling Cycles
2
Spring 2003CS 4612 Observations Emerging commodity components can be used to build IP routers –switching fabrics, network processors, … Routers are being asked to support a growing array of services –firewalls, proxies, p2p nets, overlays,...
3
Spring 2003CS 4613 Data Plane (IP) Control Plane (BGP, RSVP…) Router Architecture
4
Spring 2003CS 4614 Software-Based Router + Cost + Programmability – Performance (~300 Kpps) – Robustness Data Plane (IP) Control Plane (BGP, RSVP…) PC
5
Spring 2003CS 4615 Hardware-Based Router – Cost – Programmability + Performance (25+ Mpps) + Robustness Data Plane (IP) Control Plane (BGP, RSVP…) ASIC PC
6
Spring 2003CS 4616 NP-Based Router Architecture + Cost ($1500) + Programmability ? Performance ? Robustness Data Plane ( packet flows) Control Plane ( packet flows) IXP1200 PC
7
Spring 2003CS 4617 In General... IXP Pentium IXP Pentium... Pentium
8
Spring 2003CS 4618 Architectural Overview... Network Services... Virtual Router... Hardware Configurations... Packet Flows Forwarding Paths Switching Paths
9
Spring 2003CS 4619 Virtual Router Classifiers Schedulers Forwarders
10
Spring 2003CS 46110 Simple Example IP Proxy Active Protocol
11
Spring 2003CS 46111 Intel IXP ScratchDRAMSRAM 6 Micro- Engines StrongARM FIFOs IX Bus MAC Ports IXP1200 Chip PCI Bus
12
Spring 2003CS 46112 Processor Hierarchy MicroEngines Pentium StrongArm
13
Spring 2003CS 46113 Data Plane Pipeline DRAM (buffers) SRAM (queues, state) Input FIFO Slots Output FIFO Slots Input Contexts Output Contexts 64B
14
Spring 2003CS 46114 Data Plane Processing INPUT context loop wait_for_data copy in_fifo regs Basic_IP_processing copy regs DRAM if (last_fragment) enqueue SRAM OUTPUT context loop if (need_data) select_queue dequeue SRAM copy DRAM out_fifo
15
Spring 2003CS 46115 Pipeline Evaluation 100Mbps Ether 0.142Mpps Measured independently
16
Spring 2003CS 46116 What We Measured Static context assignment –16 input / 8 output Infinite offered load 64-byte (minimum-sized) IP packets Three different queuing disciplines
17
Spring 2003CS 46117 Single Protected Queue Lock synchronization Max 3.47 Mpps Contention lower bound 1.67 Mpps O I I I Output FIFO
18
Spring 2003CS 46118 Multiple Private Queues Output must select queue Max 3.29 Mpps O I I I Output FIFO
19
Spring 2003CS 46119 Multiple Protected Queues Output must select queue Some QoS scheduling (16 priority levels) Max 3.29 Mpps O I I I Output FIFO
20
Spring 2003CS 46120 Data Plane Processing INPUT context loop wait_for_data copy in_fifo regs Basic_IP_processing copy regs DRAM if (last_fragment) enqueue SRAM OUTPUT context loop if (need_data) select_queue dequeue SRAM copy DRAM out_fifo
21
Spring 2003CS 46121 Cycles to Waste INPUT context loop wait_for_data copy in_fifo regs Basic_IP_processing nop … nop copy regs DRAM if (last_fragment) enqueue SRAM OUTPUT context loop if (need_data) select_queue dequeue SRAM copy DRAM out_fifo
22
Spring 2003CS 46122 How Many “NOPs” Possible? IXP1200 Evalualtion Board 1.2 Mpps = 8x100Mbps
23
Spring 2003CS 46123 Data Plane Extensions
24
Spring 2003CS 46124 Control and Data Plane Smart Dropper Layered Video Analysis (control plane) (data plane) Shared State
25
Spring 2003CS 46125 What About the StrongARM? Shares memory bus with MicroEngines –must respect resource budget What we do –control IXP1200 Pentium DMA –control MicroEngines What might be possible –anything within budget –exploit instruction and data caches We recommend against –running Linux
26
Spring 2003CS 46126 Performance MicroEngines Pentium StrongArm 310Kpps with 1510 cycles/packet 3.47Mpps w/ no VRP or 1.13Mpps w/ VRP buget
27
Spring 2003CS 46127 Pentium Runs protocols in the control plane –e.g., BGP, OSPF, RSVP Run other router extensions –e.g., proxies, active protocols, overlays Implementation –runs Scout OS + Linux IXP driver –CPU scheduler is key
28
Spring 2003CS 46128 Processes........................ Input Port Pentium Output Port P P PP P
29
Spring 2003CS 46129 Performance
30
Spring 2003CS 46130 Performance (cont) Kpps
31
Spring 2003CS 46131 Scheduling Mechanism Proportional share forms the base –each process reserves a cycle rate –provides isolation between processes –unused capacity fairly distributed Eligibility –a process receives its share only when its source queue is not empty and sink queue is not full Batching –to minimize context switch overhead
32
Spring 2003CS 46132 Share Assignment QoS Flows –assume link rate is given, derive cycle rate –conservative rate to input process –keep batching level low Best Effort Flows –may be influenced by admin policy –use shares to balance system (avoid livelock) –keep batching level high
33
Spring 2003CS 46133 Experiment A (BE) B (QoS) C (QoS) A + C B
34
Spring 2003CS 46134 Mixing Best Effort and QoS Increase offered load from A
35
Spring 2003CS 46135 CPU vs Link Fix A at 50Kpps, increase its processing cost
36
Spring 2003CS 46136 Turn Batching Off CPU efficiency: 66.2%
37
Spring 2003CS 46137 Enforce Time Slice CPU efficiency: 81.6% (30us quantum)
38
Spring 2003CS 46138 Batching Throttle Scheduler Granularity: G –flow processes as many packets as possible w/in G Efficiency Index: E, Overhead Threshold: T –keep the overhead under T%, then 1 / (1+T) < E Batch Threshold: B i –don’t consider Flow i active until it has accumulated at least B i packets, where C sw / (B i x C i ) < T Delay Threshold: D i –consider a flow that has waited D i active
39
Spring 2003CS 46139 Dynamic Control Flow specifies delay requirement D Measure context switch overhead offline Record average flow runtime Set E based on workload Calculate batch-level B for flow
40
Spring 2003CS 46140 Packet Trace
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.