Presentation is loading. Please wait.

Presentation is loading. Please wait.

Programmable Networks: Active Networks + SDN. How to Introduce new services Overlays: user can introduce what-ever – Ignores physical network  perf overhead.

Similar presentations


Presentation on theme: "Programmable Networks: Active Networks + SDN. How to Introduce new services Overlays: user can introduce what-ever – Ignores physical network  perf overhead."— Presentation transcript:

1 Programmable Networks: Active Networks + SDN

2 How to Introduce new services Overlays: user can introduce what-ever – Ignores physical network  perf overhead – Overlay nodes – software routing  perf overhead Middleboxes: ops can introduce what-ever – Must be placed in a specific location – Must determine apriori what type of MB you want

3 In the ideal work Ideal Anyone can introduce services – So, like overlay Can achieve Data-plane throughput – So, like MB Can introduce anything – So like Overlay Problem: – How to run untrusted code in your environment – We want flexible but need to deal with security, performance, safety

4 Enter Active Networks Motivated by advances in RPC Goal: run mobile code in network – Code  implementation of a new service Active Extensions: User ships code to network devices – All packets use the code – No change to packet format Active Packets: Each packet carries the code or pointer to the code – Very fine-grain control – Packet is larger – A lot of redundant data

5 REALITY, MERGE BOTH: The How? Entities are a set of pre-installed modules. The Active packet include the graph of which entities to use. Active Packet! Type:Serv 1 Type:Serv5 Type:Serv6 Type:Serv 1 Type:Serv5 Type:Serv6 IP Header TCP Header Network Router Serv 1 Serv 2 Serv 5 Serv 6 Serv 7

6 REALITY, MERGE BOTH: The Why? Promotes more modularity and reuse – Entities can be smaller; since packet can be used to stitch together multiple entities. Reduces waste of N/W – A flow  multiple packets – If each Packet carries code  lot of n/w wasted Lots of Waste CPU, load/unloading code Wasted n/w b/w because of code in pkt

7 REALITY, MERGE BOTH: End-to-End Picture Network Router Active Packet! Type: serv1 Serv 1 Global Entity Store Serv 1 Serv 2 Serv 3 Signs code with special Key, so routers know to trust the code The type is an MD5 hash Of the code. This way pkts are treated by the exact code you downloaded from store

8 REALITY, MERGE BOTH: End-to-End Picture Network Router Active Packet! Type: serv1 Serv 1

9 REALITY, MERGE BOTH: End-to-End Picture Network Router Active Packet! Type: serv1 Serv 1

10 REALITY, MERGE BOTH: End-to-End Picture Network Router Active Packet! Type: serv1 Serv 1

11 REALITY, MERGE BOTH: End-to-End Picture Network Router Active Packet! Type: serv1 Serv 1 Each router caches the code so that it can be used for Next packet. Packet only caries a pointer to the code.

12 Active Packet: Capsules Recall: OSI  layering – A.P.  just random modules no need to stick to layers

13 Routers: Active Nodes A VM (JVM? Language level safety) – Protect code from each other – Prevent for interfering with each other A Trusted Operating system – Allow sharing of resources – Need Some that interfaces directly with H/W Network Router Linux OS JVM Serv 1

14 Challenges: Performance Traffic must be similar + bursty – Or else caching wouldn’t work Network has diff types of nodes – Not all can run code at line rate Think: Core V Edge Only run on edge nodes

15 What impacts Performance of Node

16 Code distribution – Caching of code minimizes this Random management tasks – Cleanup memory (GBC), run normal protocols Running code Main performance bottle-neck

17 Challenges: Security one code changing with another code's state – No sharing of state due to sandbox. Node O.S. maybe corrupted by code – Sandbox prevents this. Sending bad/malicious code to a node – The person signing should catch bad code – (Think Apple’s App store) Pkt/Capsule using the wrong code at node – Wrong code will have diff finger print, – so finger print in pkt would make finger print of code at node Network Router Linux OS JVM Serv 1 10001 00110 01000 10001 10111 01000 Good Serv 1 Good Serv 1 Bad Serv 1 Bad Serv 1

18 Challenges: Resource Sharing code using too much resource on a node – Limit resource consumption (also limit code size) – Kill code if it runs for too long code using too much resource across a set of nodes: Tricky – – Use TTL to prevents loops. If I make copies – then they all get the same TTL – Divide TTL when making copies Doesn't work for multicast. an app sending too many capsules/pkts – Similiar to today's internet.

19 Limitation of API Fixed assumption that code must work around: – 1. format of IP – 2. resource limits (TTL & size & time) – 3. code distribution – 4. how code types are computed and calculated

20 Limitation in terms of Architecture Things that can't be easily specified: – FW --> since it should work for all flows just not the flows with the type specified. – Web-Cache/transcoders --> code is short lived.

21 Why this Never took off? Performance. – Still relatively slow– only at edge Complex changes to routers – Routers should run JVM Only a few types of networks – ISP and maybe Enterprise networks – So very limited use-cases

22 A New Problem

23 Operator’s Goal Network Reality Interface vlan901 ip address 10.1.1.5 255.0.0.0 ospf cost 100 ip access-group 9 out ! Router ospf 1 router-id 10.1.2.23 network 10.0.0.0 0.255.255.255 ! access-list 9 10.1.0.0 0.0.255.255 23

24 Old Solution: Programmable Networks Ethane Sw1 Sw2 Sw3 Packet

25 Ethane Drawbacks Require complex hardware – Each switch needs to encrypt/decrypt packets Performance issues – The controller is involved with every packet

26 Practical Solution: SDN (e.g. OpenFlow) If (port == 80) Then Drop If (port == 22) Then send on if 2

27 OpenFlow API (0.9) Match – IP – Mac – Port – VLAN – TOS Action: – Forward/flood on specific interfaces – Drop packet – Rewrite ip or mac headers Layer 3: (OSPF) 1. Matches on IP address 2. Forwards on a port Layer 3: (OSPF) 1. Matches on IP address 2. Forwards on a port Layer 2: (Spanning Tree) 1. Matches on MAC address 2. Forwards on a port OR 2. Floods the packet Layer 2: (Spanning Tree) 1. Matches on MAC address 2. Forwards on a port OR 2. Floods the packet Layer 3.5: (Firewall/ACL) 1.Matches on IP address OR 1.Matches on a port 2.Drops or forwards the pkt Layer 3.5: (Firewall/ACL) 1.Matches on IP address OR 1.Matches on a port 2.Drops or forwards the pkt Layer 2.5: (Spanning Tree) 1.Matches on VLAN 2.2. Floods the packet Layer 2.5: (Spanning Tree) 1.Matches on VLAN 2.2. Floods the packet

28 OpenFlow API Layer 3: (OSPF) 1. Matches on IP address 2. Forwards on a port Layer 3: (OSPF) 1. Matches on IP address 2. Forwards on a port Layer 2: (Spanning Tree) 1. Matches on MAC address 2. Forwards on a port OR 2. Floods the packet Layer 2: (Spanning Tree) 1. Matches on MAC address 2. Forwards on a port OR 2. Floods the packet Layer 3.5: (Firewall/ACL) 1.Matches on IP address OR 1.Matches on a port 2.Drops or forwards the pkt Layer 3.5: (Firewall/ACL) 1.Matches on IP address OR 1.Matches on a port 2.Drops or forwards the pkt Layer 2.5: (VLAN) 1.Matches on VLAN 2.2. Floods the packet Layer 2.5: (VLAN) 1.Matches on VLAN 2.2. Floods the packet HP ACL OSPF VLAN SPT HP Magic Protocols Cisco ACL OSPF VLAN SPT Cisco Magic Protocols Juniper ACL OSPF VLAN SPT Juniper Magic Protocols

29 OpenFlow API Simple Firmware patch HP ACL OSPF VLAN SPT HP Magic Protocols Cisco ACL OSPF VLAN SPT Cisco Magic Protocols Juniper ACL OSPF VLAN SPT Juniper Magic Protocols OpenFlow Switch ACL OSPF VLAN SPT

30 Lesson A rigid network is impractical – Doesn’t support new services Programmable Networks allow great flexibility – Allows anyone to introduce new services – Into which ever nodes they have access to But this flexibility introduces new challenges – Security, performance, Resource control For Technological adoption – Minimal overhead for transition is good – New h/w is hard to get created


Download ppt "Programmable Networks: Active Networks + SDN. How to Introduce new services Overlays: user can introduce what-ever – Ignores physical network  perf overhead."

Similar presentations


Ads by Google