Upload
linaro
View
333
Download
2
Tags:
Embed Size (px)
DESCRIPTION
Resource: LCU13 Name: High Performance Networking - Deep dive Date: 29-10-2013 Speaker: Jarmo Hillo / Raj Murali Srinivasan Video: http://www.youtube.com/watch?v=QTtKOoIteaY
Citation preview
©2013 Nokia Solutions and Networks. All rights reserved.
High end telecom networking – deep dive
LCU 2013
2 ©2013 Nokia Solutions and Networks. All rights reserved.
1000x Packet Traffic Technologies are Maturing
… times more capacity
3D silicon
process for
maximum
efficiency in
minimum
space
Optimal
architectural
split for variety
of payload
Ultrafast
interfaces
reaching
100Gbit/s
3 ©2013 Nokia Solutions and Networks. All rights reserved.
Single core networking
Only priority queuing doable with
single core
• Static priorities
• Highest priority first with
starving prevention logic
• Packet order not an issue
• Does not scale
Classify
100%
4 ©2013 Nokia Solutions and Networks. All rights reserved.
Multicore SoC Load Sharing
Multicore SoC great for load
sharing • Static resource allocation
• Queue selection typically by hash of incoming
packet header or round robin
• HW aware applications
• Unpredictable, uneven core load
Hash
95% 5% 60% >100%
5 ©2013 Nokia Solutions and Networks. All rights reserved.
Multicore SoC Load Balancing
Multicore SoC great for load
balancing • Dynamic resource allocation with rules
• Fast per packet decision
• Event Machine model: each thread asks for
new job after current is finished
• Automatic scaling – “single thread”
programming model
Scheduler
65% 65% 65% 65%
6 ©2013 Nokia Solutions and Networks. All rights reserved.
Hickups with multicore scaling
Relative speedup does not
scale linearly with cores, for
example 64 cores, 1% serial
results to less than 40x speedup
7 ©2013 Nokia Solutions and Networks. All rights reserved.
• Mobile traffic for large amount of cells
• Find max throughput - QoS threshold for adding more users
Case study: WCDMA HSPA
Static core allocation • Users allocated per core based on resource
availability to maintain enough peak reserves
• High headroom
required
• Low average load
Dynamic core allocation • Allocation of users to a resource pool of cores
• Statistical multiplexing gain
• Low headroom
• High average load
Core1 Core2 Core3 Core4
Avera
ge lo
ad
Peak r
eserv
e
Core5
~50%
Core1 Core2 Core3 Core4 Core5
~90%
80% gain
8 ©2013 Nokia Solutions and Networks. All rights reserved.
Case study: Linux userspace PMD vs. socket
packet test
HW load balancer further improves scaling with number of cores
Dataplane in userspace scales linearly
Kernel approach saturates at ten cores
9 ©2013 Nokia Solutions and Networks. All rights reserved.
HW abstraction of networking SoC
Logical functionality and packet
flow in typical networking SoC
Null latency intelligent packet scheduling to the cores
10 ©2013 Nokia Solutions and Networks. All rights reserved.
SW abstraction of networking SoC Open Event Machine SW architecture
11 ©2013 Nokia Solutions and Networks. All rights reserved.
Abstraction
Scalability
Efficiency
What do we need
1 x
12 ©2013 Nokia Solutions and Networks. All rights reserved.
No need to (re)program as SoC or core resources increase. “Single core” programming model
Maintain same architectural split across platforms
Agnostic to programming model
Application portability
Common terminology
Abstraction is the key for portability
Resources
API
Functional
SoC
core core core …
Applications
13 ©2013 Nokia Solutions and Networks. All rights reserved.
Optimal balance of cores, accelerators and interfaces
Eliminates single core bottleneck
Allows dynamic load balancing
Minimize the serialization
Ready for ultrafast interface technologies
No packet drop before classification
Scalability for optimal capacity
HW offload
Interfaces
Optimal number of resources
14 ©2013 Nokia Solutions and Networks. All rights reserved.
User space direct HW access
No context switching
Idle time power saving
Moore’s Law, FinFet
Dedicated processing units
Efficiency for maximum density
Power efficiency
Compute efficiency
22nm 7nm
15 ©2013 Nokia Solutions and Networks. All rights reserved.
Open Data Plane
OpenDataPlane™
ODP will be the de-facto data plane programming model
Common API for networking SoC
Abstraction with performance
Application portability across platforms, time to market
Scaling to any number of cores and accelerators
Efficiency in mind
16 ©2013 Nokia Solutions and Networks. All rights reserved.
How does ODP map to Openstack
ODP
ODP
17 ©2013 Nokia Solutions and Networks. All rights reserved.
How does ODP map to NFV
ODP ODP ODP
18 ©2013 Nokia Solutions and Networks. All rights reserved.
How does ODP map to Open Daylight
ODP
ODP ODP ODP
ODP
19 ©2013 Nokia Solutions and Networks. All rights reserved.
Resource and
functional
abstraction
Scalablility
for variety of
payload
Compute
efficiency
Summary
… get ready for 1000x packet compute
©2013 Nokia Solutions and Networks. All rights reserved.
1GB of personalized
data per user per day
means 1000x capacity
compared today.
Questions?