QAD System Monitoring for Efficient Operations Derek Bradley – Performance Engineer, Architect, QAD Managing and Monitoring QAD Systems

QAD System Monitoring for Efficient Operations

Derek Bradley – Performance Engineer, Architect, QAD

Managing and Monitoring QAD Systems

2

The following is intended to outline QAD’s general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, functional capabilities, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functional capabilities described for QAD’s products remains at the sole discretion of QAD.

Safe Harbor Statement


3

• This breakout session introduces best practices in managing and monitoring QAD systems – and how QAD can help

• Managing the technical infrastructure of complex computer systems is a challenge

• The needs and expectations of the users need to be balanced against the resources and requirements of the IT management team

Introduction


4

• Database administrators• Systems administrators• Technical project managers• IT management• Survey

Aim / Audience


5

• Keep the users and sponsors of the QAD software installation happy and productive- Up and down the supply chain

• Reliability• Performance• Visibility• Reporting

Best Practice Outcomes


6

Reliability


7

• Use Highly Available Architecture- Along with well trained staff and good management systems

• HA systems employ fault tolerance, automated failure detection, recovery, testing, problem and change management

• Duplicate everything. - Eliminate single points of failure

Managing and Monitoring QAD Systems: Reliability

8

• High Availability Systems have the following technical design requirements- Heartbeat monitoring- Scripts or tools to start / stop / failover and

failback the QAD application- Shared storage (SAN)- Non−Corruption of data when the failover

occurs• After imaging / replication

High Availability Guidelines


9

Disaster Recovery Planning


• Disaster recovery planning (DRP) is essential for any company

• Of companies that have had a major loss of business data- approximately 40% never re-open- 50% close within 2 years- fewer than 10% survive long term

(source : Hoffer, Jim. "Backing Up Business - Industry Trend or Event." Health Management Technology, Jan 2001).

10

• Tier 0 : No offsite data

• Tier 1: Offsite backup but no “hot” site

• Tier 2: Offsite backup and “hot” site

• Tier 3: Electronic vaulting

• Tier 4: Point in time copies

- After imaging, disk flash copy

• Tier 5: transaction integrity

- OE replication, disk replication

• Tier 6: zero data loss

• Tier 7: completely automated

The Seven Tiers of Disaster Recovery


11

Disaster Recovery Objectives


• Working with the business, decide upon the following objectives - Recovery Time Objective (RTO)

• How long can you afford to be without your systems?

- Recovery Point Objective (RPO)• When it is recovered, how much data can you afford

to recreate?

- Degraded Operations Objective (DOO)• What will be the impact on operations with fewer

data centers?

- Network Recovery Objective (NRO)• How long to switch over the network?

12


Performance

13

A Commonly Misunderstood Subject: What is Performance?

Managing and Monitoring QAD Systems: Performance

Responsiveness?

High Capacity?

Low System Requirements?

14

• Users and IT support = application responsiveness

• Database administrators = database efficiency and potential bottlenecks

• Systems administrators and engineers = server capacity and utilization

• IT managers = user productivity, system availability, budgets and risk avoidance

Roles and Responsibilities


15

• Slow and unresponsive applications• Unexplained / random application freezes• Batch processes fail to complete (in time)• Lack of scalability• Lack of capacity

Typical Performance Problems


16

Forrester Research has reported that among companies with revenue > $100 billion, nearly 85% reported significant application performance

degradation

Best Practices in Problem Management

Nearly 85% of applications are failing to meet and sustain their performance requirements over

time and under increasing load

Impact of Poor Performance


17

• Lost productivity• Lost confidence and credibility– Customers– End users

• Lost revenue• Low morale• Financial penalties

Impact of Poor Performance


18

Performance Engineering Methodology


19

• Establish performance objectives• Identify critical requirements• Define abnormal and normal conditions– Service level agreements

• Create a baseline• Continuous monitoring and alerting– QAD monitoring framework

• Performance tuning• Capacity planning and re-sizing

How Do We Manage Performance?


20

• Unless performance is actively managed and benchmarked, user performance expectations are hard to quantify.

“The system is running slow.”“It takes too long to log in.”

What do these mean? Can we determine critical / objective requirements?

Performance Objectives


21

• Using KPIs and performance requirements– Create a set of baseline measurements– Capacity requirements planning &trending

• Load testing tools may help with creating a baseline– Apache Jmeter– HP LoadRunner

• QAD Monitoring

Establish a Baseline


22

QAD Monitoring


23

QAD Monitoring

Monitoring QAD Systems

24

Ad Hoc Monitoring lacks transparency and can lead to emergency performance escalations

• Continuous monitoring allows- Advanced notice of developing problems- Trending against the baseline- Extra information to aid in problem solving- The ability to deliver KPI information to

management on demand

QAD Monitoring


25

• Holistic system monitoring• Visual correlation of data• Visibility into system trends and usage• Ability to deliver KPI information on

demand• Powerful warning and exception alerting• Reporting framework• Scalable / flexible

Key Features


26

Technical Features


• Non−intrusive, lightweight monitoring• Technology agnostic

- Can monitor any component in the QAD technology stack on any platform

- Any supported database technology

• In built wiki with full documentation• Industry standard open source

components- Proprietary QAD integration pieces

27

Graphing and Trending Features


• Allows graphing of numerical data for trending and analysis- Helps identify usage patterns and trends

• Enables visual correlation of data

- Data stored in time series (RRD) database- Filter by time periods of 30m to 1 year- Template driven deployment

28

• Whenever a pre-defined condition is met, an alert can be sent to one or more contacts- Email / Pager / Twitter / Phone / Chat

- Warning, critical and unknown alert levels• Recovery messages

- Time zones, rosters• Escalation paths

- Template driven definitions• Inheritance and overrides

- Stores service level agreement data for reporting

Alerting Features


29

Reporting and Service Level Agreements (SLA)


30

Mobile Device Support


* 3rd party apps ** webapp on appliance

31

Technolo

gy

Architecture

Technology & Architecture


32

• Deployed as a virtual appliance (Linux VMware image) - No Open Edge or commercially licensed

components installed- ESX ready or VMware Server ready version

• Security to monitored systems- Communicates with the monitored servers via

trusted SSH relationships- Keys are stored on the VM and pushed to the

remote servers

Deployment


33

• Technology agnostic:- it does not care what is being monitoring- the flexibility to monitor practically anything- version independent

• Integration templates for:- mapping to the QAD Architecture- Tomcat, Open Edge / other databases,

Connection Manager, QAD business logic- any supported Operating System*

- (Windows Support Currently Limited)

Integration


34

• On Demand customers• Now• 75 servers, 90+ environments, 2500+ services

• Early adopters• June 2011

• General Release• September 2011• Staged rollout

Availability


35

Availability


• QAD Monitoring will be available to customers who - Are on current QAD Maintenance- Who agree to a technical Q-Scan

• Establish baseline system health• Tailor QAD Monitoring to their needs• Training

37

• Continued return on existing investment• Faster response times• Higher system availability• Problem avoidance

Metrics Affected


38

• Stop by the EXPO to discuss what QAD System Monitoring can do for your company

• Ask your account manager to arrange a Q−Scan

Next Steps


39

• Tony Winter (Chief Technology Officer): [email protected]

• Paul Newton (Project Manager): [email protected]

• Derek Bradley (Architect/Consultant): [email protected]

Questions & Answers


mailto:[email protected]



40

www.qad.com© QAD Inc

http://www.qad.com/

Documents

QAD System Monitoring for Efficient Operations Derek Bradley – Performance Engineer, Architect, QAD Managing and Monitoring QAD Systems