Reliability and availability in SOA

1. Arnon
Rotem-Gal-Oz
Building reliable systems
from unreliable components

2. Arnon Rotem-Gal-Oz

3. What’s in a 9

4. 0.99 reliability
We have a nice little legacy
business component

5. And we move it to SOA
0.99 0.99 0.99
0.99 0.99 0.99
0.99 0.99 0.99

6. Failsafe hardware
Status Technologies
FT Server

7. Or try to detect failure , handle it and
minimize its effect on the business service
© Rosendahl

8. BEFORE WE BEGIN

9. SOA
Adheres to Policy governed by
Binds to End Point Exposes
Serves
Service
Understands Contracts implements Service
Consumer
describes
Key
Component
Sends/Receives Messages Sends/Receives Relation

10. Distributed Pipes and Layered Client
Agents Filters System Server
Stateless
Comm.
SOA is derived from
SOA other styles

11. SOA vs. REST
Distributed Pipes and Layered Client Replicated Uniform Virtual
Agents Filters System Server Repository Interface Machine
Stateless Code On
Cacheable
Comm. Demand
SOA REST

13. System Mobile Integration
3G Video Dedicated
Monitoring MMS 3rd parties
Calls Client
Usage
Datmart Applications Ad Management
Targeted Campaign
Billing Acquisition Interactions branding
Advertizing Mgmt.
Reporting Resources
Data mining Reference
& Statistics Interactions Links
Data
Link Managment
Reports
Publishing
Data Web Interaction
tools
Interfaces Front-end Designer
integration

14. Load balancer Smart
phones
Web Web MMS 3G 3G
Server Server Gateway Gateway Gateway
(IIS/Apache (IIS/Apache Camera
Phones
DMZ
Firewall
Load balancer
App App App App App
App App
Server Server Server Server Server
Server Server
Operational
Firewall
Backend
BI & Sync. Paper
NMS Links
Reporting Registeration Server Editor
Web
Firewall
Server
(IIS/Apache
DB Datamart DB Advertizing
clients
Links usage References Admin
Console
DMZ

15. CHALLENGE – SERVICE
AVAILABILITY

16. What’s the effect of a failure - Server
E1 line = 30
concurrent video
Call Flow
calls
Service

17. Service Instance
Service Business
logic
End point
reaction Distribute
request Dispatcher
Edge
Service Instance

18. What’s the effect of a failure - Channel
Call Flow
Call Flow Channel Call Flow
Call Flow
Call Flow Call FlowChannel
Channel
Channel Flow Channel
Call
Channel
Call Flow
Channel
Call Flow Call Flow
E1 line = 30 Channel Call Flow
concurrent video Call Flow Channel Channel
Channel
calls Channel Call Flow
Call Flow Channel FlowCall
Channel Call FlowChannel
Call FlowChannel
Channel

19. Service Instance with NLB
Real IP : 1.1.1.4 Real IP : 1.1.1.3 Real IP : 1.1.1.2
Windows Host Cluster Host Cluster Host
Service Service
Edge Instance
Instance
Windows Kernel Windows Kernel Windows Kernel
TCP/IP TCP/IP
TCP/IP
NLB Driver NLB Driver
NIC Driver
NIC Driver NIC Driver
NIC NIC NIC
Virtual IP : 1.1.1.1

20. Virtual Endpoint

21. Request/Reply
EndPoint
2.
1. Request
Synchronous
processing
3. Reply
Service Consumer Service

22. Things look Cool & Simple ™
3G Session
Call negations
Render Image
resutls extraction
Translation
identification
to links

23. 3G GW
IVP (RV)
(RV) SIP Listner
3G VAS (Cestel)
RTP Image Extractor
WS 3G Builder (Cestel)
Alg. Engine
WebRenderer
Resource Manager
Dispatcher
WebConnector
Turn out Complicated & Ugly ™

24. Parallel Pipelines
Key
SOA Component Pattern Component Queue
Relation Concern/attribute
pipeline pipeline
Edge
Request EndPoint
Request 1 EndPoint
EndPoint
Request 2 Perform Perform
Reaction Task Task
pipeline
EndPoint
Perform
Task Service

25. Inversion of Communications

26. Consumer view
var sendMmsEvent = new SendMmsEvent()
{
FromNumber = simpleMessageDetails.DialedNumber,
Subject = mmsContents.Subject,
ToNumber = simpleMessageDetails.Sender,
ImageExtension = mmsContents.ImageExtension,
ImageAsByteArray = mmsContents.Image,
TextAsByteArray = mmsContents.Text
};
eventBroker.RaiseEvent(sendMmsEvent);
http://www.flickr.com/photos/crimson_wolf/2851737125/sizes/l/

27. Service view
[ServiceContract]
public interface ImPostOffice : ImContract, IHandleSendCoupon,
IHandleSendSms,IHandleStatus,IHandleAdminStatus,IHandleWapLink,
IHandleSendMms
{
}
[ServiceContract]
public interface IHandleSendMms
{
[OperationContract]
int SendMms(SendMmsEvent eventOccured);
}
[DataContract]
public class SendMmsEvent : ImEvent
{
/// <summary>
/// end user's number. should be in international format: +[country-code]number. Example: +491737692260
/// </summary>
[DataMember]
public string ToNumber { get; set; }
/// <summary>
/// service's number, usually a short-code. Example: 84343
/// </summary>
[DataMember]
public string FromNumber { get; set; }

28. Edge translates external structures to
internal ones
public int SendMms(SendMmsEvent eventOccured)
{
var eventContext = eventOccured.ToString();
if (log.IsDebugEnabled)
log.Debug("inside 'SendMms', event context = [" + eventContext + "]");
var fromNumber = eventOccured.FromNumber;
var sender = mmsSenderFactory.Get(fromNumber);
if (null == sender)
{
if (log.IsWarnEnabled)
log.Warn("cannot get mms sender derived from '" + (fromNumber ?? "null") + "'");
return 0;
}
IMmsSubmitResponse response;
try
{
var mmsMessageDetails = new MmsMessageDetails(eventOccured.ToNumber,
eventOccured.TextAsByteArray,
eventOccured.ImageAsByteArray,
eventOccured.ImageExtension,
eventOccured.Subject);
response = sender.Submit(mmsMessageDetails);
}
catch (Exception ex)
{
log.Error("cannot send mms message, context = [" + eventContext + "]", ex);
return 0;
}
if (log.IsInfoEnabled)
{
var responseMessage = (null == response) ? "null" : response.ToString();
log.Info("sent mms with event context = [" + eventContext + "], response = [" +
responseMessage + "]");
}

29. Sagas tie instances together for
conversations

30. Call Recovery
Image
Extractor
SIP
Identification

31. Alternative : Orchestration
Service Service
reaction request
Manage Protocol
Process
Schedule Coordinator
route
request monitor
Workflow instance Offline
designer
Host
Workflow Engine Workflows Auxiliary tools
Orchestration platform

32. Be Wary of Nano-Services

33. CHALLENGE - MANAGEMENT

34. Blogjecting Watchdog
Edge Service
EndPoint
Monitor
Watchdog
Watchdog
Monitor Agent
Request Edge
EndPoint Heal
Reports Monitor
Report
Monitor
Log Monitor

35. Blogjects concept is about
collaborating objects

36. WatchDog
Service A Service B WDWatcher

37. WatchDog
Call Recovery
Image
Extractor
SIP
Identification

39. Service Monitor
Collect
Metrics
collection Status
Fault
Monitoring Reporting &
Security Dashboarding Edge/Service
monitoring
Status
Policy
governance Notify
Commands
Control In
Monitor Act Edge/Service
Service Monitor

40. RESTful resource management
root 1 2 3
Abcde/
Sessions/
Efgh/
/
Resources/ Dispatchers/ Xyz/

41. http://devrig:52141/RM/Sessions/abc/
• ATOMPUB
– Session details
• URI (ID)
• State (start/end/status etc.)
• Resources
– Knows status
– URI for the Resource representation on the RM
– URI for the Resource itself

42. Keep the BIT
http://www.flickr.com/photos/37643027@N00/2050024263/sizes/o/

43. WatchDog
Call
Recovery
Image
Extractor
X
SIP
Identification
WatchDog
3G Call
Liveliness
Monitor

44. CHALLENGE – MULTI-TENNANCY
(LIES, I TELL YOU, ALL LIES)

45. Same event different subscribers
Player
(interaction Call Flow
Renderer)
Play Movie
Event
Bridge to 3rd
Call Flow
Party

46. Routing
[ServiceContract]
[Participate("3G")]
public interface ImPlayer : ImContract, IHandleCallStarted, IHandleC
IHandlePlayMovie,IHandleCallAborted
{
}
[ServiceContract]
[Participate("3GPartner")]
public interface ImXsightsGateWay : ImContract,
IHandleCallAborted,
IHandlePlayMovie,
IHandleReadyForSearch,
IHandleSearchStarted,
IHandleJoinThirdParty
{
}

47. Raise a saga initiating event
Initiator A Participant A
Capacity : 2
Participant B
Initiator B Capacity : 1
Now what ?!

48. Reservation Pattern

49. Service Instance #1
Control Control
Edge Edge Service Instance #2
Business Event Control Business
Event
Logic Broker Edge
Broker Logic
Resource Event Business
Resource
Allocator Service Host Logic
Broker
Allocator
Service Host
Resource Host
Service
Service Host
Allocator
Service Host Service Host
Control
Reservation Edge
Event Business
Broker Logic
Resource
Service Host
Allocator
Service Host

50. Good old
2PC to the
rescue

51. Takeaways
• Things break
• Decouple
• Fail Fast
• Monitor & Detect
• Compensate
• Throw state unto others

52. Arnon Rotem-Gal-Oz
@arnonrgo
http://arnon.me
http://www.cloudvalue.com