AIOps and IT Analytics at the Crossroads: What’s Real Today and What’s Needed for Tomorrow

IT & DATA MANAGEMENT RESEARCH,
INDUSTRY ANALYSIS & CONSULTING
Dennis Drogseth
Vice President
Enterprise Management Associates
AIOps and IT Analytics at the
Crossroads: What’s Real
Today, and What’s Most
Needed for Tomorrow?

Watch the On-Demand Webinar
2
• AIOps and IT Analytics at the Crossroads: What’s
Real Today and What’s Needed for Tomorrow On-
Demand webinar is available here:
https://ema.wistia.com/medias/o6j5g2wgxf
• Check out upcoming webinars from EMA here:
http://www.enterprisemanagement.com/freeResearch

INDUSTRY ANALYSIS & CONSULTING3
Dennis Nils Drogseth, Vice President, EMA
Dennis joined Enterprise Management Associates in 1998 and currently
manages the New Hampshire office. Dennis brings several years of
experience in various aspects of marketing and business planning for
service management solutions. He supports EMA through leadership in IT
Service Management (ITSM), CMDB systems, as well as megatrends like
advanced operations analytics, cross-domain automation systems, IT-to-
business alignment, and service-centric financial optimization. Dennis also
works over several practice areas to promote dialogue across critical areas
of technology and market interdependencies.
Featured Speaker

Logistics
4
An archived version of the event recording
will be available at
www.enterprisemanagement.com
• Log questions in the chat panel located on
the lower left-hand corner of your screen
• Questions will be addressed during the
Q&A session of the event
QUESTIONS
EVENT RECORDING

INDUSTRY ANALYSIS & CONSULTING6 © 2018 Enterprise Management Associates, Inc.
Sponsors

INDUSTRY ANALYSIS & CONSULTINGSlide 7 © 2018 Enterprise Management Associates, Inc.
Agenda
• Demographics
• Overall analytic and use case priorities
• Organization and best practices
• Technology and design priorities
• Functional priorities, automation, AI bots
• Cloud, agile/DevOps and IoT
• Operationalizing advanced IT analytics—deployment,
roadblocks and success
• Conclusion: seven outstanding findings

Demographics

Respondent Base and Geography
300 respondents:
• 191 in North America
• 109 in Europe
Strong executive presence with 40%
VP and above
• Examined 4 groups:
• Executive (not including CISO)
31%
• Security (including CISO) 21%
• ITSM/operations 20%
• Technical support (data scientist,
data management, engineering,
etc.) 20%

Balanced Spread for Company Size: 35%
Small Enterprise; 30% Mid-Tier Enterprise;
35% Large Enterprise
0%
0%
11%
24%
17%
13%
17%
18%
Less than 250
250-499
500-999
1,000-2,499
2,500-4,999
5,000-9,999
10,000-19,999
20,000 or more
How many employees are in your company worldwide?
Less than 250
250-499
500-999
1,000-2,499
2,500-4,999
5,000-9,999
10,000-19,999
20,000 or more
Sample Size = 300

Verticals and Types of Involvement
Lead verticals:
• High tech software (ISVs) (15%)
• Technology service providers (11%)
• Manufacturing (10%)
• Finance/banking (9%)
Types of involvement
• Managerial oversight (39%)
• Hands-on stakeholder (33%)
• Technical stakeholders (data scientists, etc.) 24%
• Business stakeholders (4%)

Overall Analytic and
Use Case Priorities

Advanced IT Analytics (AIA) and
AIOps Confluence
1. Assimilation of data from cross-domain sources in high data
volumes for cross-domain insights
2. Access multiple data types, e.g., events, KPIs, logs, flow,
configuration data, etc.
3. Capabilities for self-learning to deliver predictive, and/ or
prescriptive and/or if/then actionable insights
4. Support for a wide range of advanced heuristics
5. Potential use as a strategic overlay that may assimilate
multiple monitoring investments
6. Support for private cloud and public cloud
7. The ability to support multiple use cases

EMA Quotas Targeted AIOps
65%
12%
10%
11%
2%
0%
0%
0%
AIOps across multiple domains (or IT operations analytics) (or
digital operations)
Big data stores for data search
End-user experience/customer experience management analytics
Security-specific analytics
Capacity-specific analytics
Other
Don't know
None of the above
What types of analytic investment in support of IT are YOU primarily engaged in?
Sample Size = 300

AIOps in Profile
When respondents were asked to align attributes as they
perceived them with AIOps, the top seven were:
• Dataset aggregation
• Big data analytics
• Higher levels of automation
across IT
• Machine learning
• Behavioral learning
• Intelligent incident management
• Supervised learning
Average respondent checked more than 7 (7.25) options

AIOps vs. Other AIA Examples
AIOps led in the following categories:
• An affiliation with larger
enterprises
• Active support for a broader
range of use cases
• More likely to be top-down
driven by the executive suite
• A greater affinity for applying best practices
• Dramatically broader support for third-party toolset
integrations
• Stronger support for integrated automation,
including AI bots
• The highest success rate overall

Use Case Priorities

Organization and
Best Practices

Executive Leadership Is
Clearly Dominant
50%
23%
16%
4%
3%
1%
2%
1%
0%
IT executive suite (CIO or VP)
Director-level IT
Manager-level IT
CISO/CSO/Chief risk or compliance officer
Chief analytics officer/chief data officer
Business executive (non-IT) line of business
VP or Director of digital business marketing/planning
VP/Director of software engineering/ development
Other
Which executive title is most likely to lead your analytics strategy?
Sample Size = 300

Stakeholders Supported –
A Total of 19 Roles
The top five domain stakeholders
(with an average of 7.21 supported)
were:
• Cloud management
• Database management
• Applications
management/support
• Security/compliance
• Systems
• The top five cross-domain stakeholders
(with an average of 7.55 supported) were:
• IT operations/cross-domain (tied with)
executive IT
• ITSM (beyond the service desk)
• Data analyst/data scientist
• Infrastructure management
• Line of business (not central IT)
• The top five business stakeholders (with an
average of 4.47 supported) were:
• Business operations
• Business development/planning
• Customer experience management
• Executive (non-IT)
• Online operations

93% Indicated Extremely or Very Good
Integration Between Operations and ITSM
51%
49%
48%
45%
43%
43%
42%
42%
41%
39%
35%
28%
28%
0%
IT governance analytics supporting operational efficiencies
Shared data for improving internal end-user experience
Active social IT support shared between users and IT
Integrated ITSM knowledgebase sharing with operations analytics
Mobile IT communications across ITSM and operations
Integrated support for SLM/SLA priorities
Integrated support for end-user experience management via analytics
Integrated project management
Support for integrated change/performance via CMDB/CMS/ADDM
Integrated trouble ticket analytics
Workflow, scheduling for triage, and remediation
Shared runbook and automation routines
Other integration between ITSM and operations for change/performance
Other
How do operations and ITSM collaborate in leveraging IT analytics?
Sample Size = 293, Valid Cases = 293, Total Mentions = 1,564

Some Perspectives on Digital
Transformation and Best Practices
94% viewed digital transformation as either an
‘extremely’ or a ‘very’ high priority, with initiatives
well under way
• An indication of success in overall AIA initiatives
• 55% see digital transformation as driving
their AIA initiatives
• And 37% see the two as tightly
coupled
63% are leveraging best practices
in support of their AIA deployments
• 35% have plans to leverage
best practices
• Best practices also correlate
with AIA success
• Top three were ISO Security 27001/27002;
Regulatory compliance (e.g. HIPAA),
IT Balanced Scorecard

Technological and
Design Priorities

The Average Response Indicate more
than Eleven (11.38) Heuristic Affinities
(Average was 3.28 in 2016)
64%
64%
60%
59%
59%
59%
58%
58%
58%
56%
56%
56%
56%
56%
55%
54%
54%
53%
52%
51%
5%
Security instrumentation
User experience analytics
Big data search, such as Qlik or Tableau
Data mining
Event analytics
Log analytics
Historical trending
Behavioral analysis
Rule-based analytics
If/then or what-if change impact analysis
Anomaly detection
Real-time predictive
Predictive modeling/emulation
Online analytical processing (OLAP) (not including data mining)
Natural language search, processing, or understanding
Machine learning
Predictive trending
Prescriptive analytics
Stream analytics
Rule correlation
Other
What type of AI-related heuristics does your organization currently use?

Data Source Priorities
Data sources showed a similar increase to an average of more
than twelve (12.65) in Q3 2018 versus five in Q1 2016. The top
five data sources in the new research were:
• Internet of Things
• Spreadsheets
• Transaction data
• Configuration/metadata
• Logfiles/access logs
The top five security-related data sources were:
• Antivirus
• Security information and event management (SIEM)
• Security log management and search
• Events/time series, security-related
• Threat intelligence

The Average Response Indicated that
AIA Investments Should Assimilate
About 23 Monitoring or Other Tools
1%
9%
14%
17%
21%
15%
8%
13%
3%
None
1-5
6-10
11-20
21-30
31-40
41-50
More than 50
Don't know
How many monitoring or other management tools would you expect to integrate into
your organizations IT analytics solutions directly or through an aggregated data store?
None 1-5 6-10 11-20 21-30 31-40 41-50 More than 50 Don't know
Sample Size = 300

Interdependencies
Top Five Interdependencies (average of
5 per respondent)
• Infrastructure-to-application
• Endpoint-to-infrastructure
• Infrastructure-to-infrastructure
• Infrastructure-to-business services
• Application-to-business services
Top Four Sources
• Application dependency mapping for
cost
• Application dependency mapping for
change
• Service modeling dashboard for
business impact
• Service modeling/topology provided
through analytic tool

CMDB/CMS Specifics
54% viewed CMDB/CMS as “extremely
important” to their AIA strategy
• 36% as “very important”
55% updated their CMDB/CMS as frequently as
under five minutes
• Real-time currency also favored success
81% are updating the CMDB/CMS-related
dependency insights via AIA, for currency and
relevance
• Which also favored success
• 17% would like to

Functional Priorities,
Automation and AI Bots

Functional Priorities: Triage, Change
Management, and Application
Infrastructure Optimization
Top three priorities for triage:
• Isolate security issues
• Isolate database issues
• Isolate issues in the network
Top three priorities for change management and
application/infrastructure optimization
• Security-related issues
• Data quality management efficiencies
• End-user experience optimization

Security, End-User-Experience and
Business Metrics
Top three security metrics
• Network detection of threats
• Relative security risk
• Fraud detection
Top three end-user-experience metrics
• Application/infrastructure performance as it impacts user
experience
• Levels of security, risk, and data integrity
• Performance of third-party components in a web service
Top three business impact metrics
• Revenue through IT services
• Business activity metrics
• Improved business efficiencies due to reduced downtime

Average Response Indicated More
Than Five (5.16) Automation Options
54%
41%
41%
40%
38%
37%
36%
35%
35%
35%
34%
34%
32%
30%
29%
1%
0%
IT process automation (and/or runbook)
Security process automation (and/or playbooks)
Workflow automation combined with social IT
Configuration automation
DevOps-related process automation
Security instrumentation (continuous attack testing and defense stack validation)
Automation in support of business-specific outcomes
Automation-driven discovery/inventory
Automation in support of data assimilation/data reconciliation
Auto-scaling/capacity optimization
Standard service desk or ITSM workflows
Advanced incident management handling (beyond trouble ticketing)
Integrated trouble ticketing
Advanced workflow integrated with automation
Alert-driven notification
None - we are not planning to use automation in support of our analytics initiatives
Other
Which types of workflow and/or other types of automation are you currently using
or planning to use in support of your analytics initiative(s)?

AI Bots
57 percent of respondents indicated that they were currently using
AI bots
• 26 percent that they had specific plans for AI bots.
Top three use cases were:
• AI bots directed at availability and performance management
• AI bots directed at managing change
more efficiently
• AI bots directed at security and
compliance concerns
44% claimed that AI bots were
tightly woven into their overall
AIA strategy
• 34% claimed that they were
somewhat integrated
• Only 2% had no plans
to integrate

Cloud, Agile/DevOps and IoT

Optimizing Hybrid Cloud and
Integrated Security and Performance
Led for AIA Use Cases vis-à-vis Cloud
19%
19%
18%
18%
17%
17%
16%
15%
14%
13%
12%
11%
10%
Hybrid cloud optimization, not including costs
Integrated security and performance
Integrated security and change
Improved storage control and cost optimization
Cloud cost optimization for on-premise/multi-cloud
Real-time service (application, etc.) performance
Overall cloud migration
Improved network security
Continuous deployment/integration (aka DevOps/agile)
Compliance
Change impact (optimizing the impacts of change)
Business impact/business outcomes
Capacity planning and optimization
What are your organizations top two (2) use cases for IT analytics in support of cloud
initiatives and cloud-related services (including hybrid cloud/non-cloud)?
Sample Size = 300, Valid Cases = 300, Total Mentions = 600

DevOps Highlights
74% were actively using AIA in support of DevOps
• Only 3% have no plans to support DevOps with AIA
• 67% see AIA and DevOps analytics as fully integrated
Top five priorities were
• Optimize application performance by providing rapid feedback to
development from production
• Minimize time developers spend troubleshooting production
performance issues
• Support the application development
process directly
• Provide feedback to optimize
application design
• Drive improvements through
end-user experience

Internet of Things (IoT) and AIA
71% were currently deploying analytics in support of IoT
• Only 3% had no plans to deploy
• 69% of the 71% viewed these as fully integrated with their
AIA/AIOps strategy
Prioritized use cases were:
• Manufacturing
• Facilities
• Utilities
• Other vertically-specific needs
• Transportation/fleets

Operationalizing Advanced IT
Analytics—Deployment,
Roadblocks and Success

Leadership, Overhead and
Roadblocks
52% were driven by the executive suite (VP and above)
The average deployment required more than 2 FTEs for ongoing
administrative support
Top five roadblocks were
• Data quality issues
• Products not fully baked yet
• Data relevance/ lack of context
• Tools are too complex to
administer
• Internal resources – getting
budget and people

Benefits and Success
Five indicators of success (and improved ROI)
• Top-down executive leadership
• Prioritizing AIOps
• More heuristics and data sources
• CMDB/CMS prioritization
• More use capabilities for triage, change management and
infrastructure optimization and business impact
Top five benefits achieved
• Improved OpEx efficiencies within IT
• Faster time to repair problems
• Faster identification of advanced
threats
• Faster time to deliver new
IT services
• Better correlation between
change and performance

Conclusion:
Seven Outstanding Findings

AIOps and AIA Eclecticism
1. AIOps was overall the winning strategy
• AIOps showed the highest success rates, the greatest likelihood of
supporting DevOps, IoT and AI bots, and led in use case
capabilities as well.
• Big data led as the most prevalent before quotas
2. Advanced IT analytics are eclectic and becoming more so
Overall support for DevOps, IoT, AI bots, and multiple use cases
including EUEM, security, capacity analytics, cost-related
optimization, show increasing diversity in need and value.

AI Bots and Security
3. AI bots are not a separate world from AIOps
and AIA
AI bot integrations indicate that the AIOps ‘market’ and the AI
bots ‘market’ should not be viewed in isolation.
4. The importance of capturing interdependencies
and the CMDB/CMS
Respondents sought to capture 4.81 interdependencies
across the application/infrastructure, while 54% of
respondents viewed the CMDB as ‘extremely important’ to
their analytics strategy.
5. Security is on the rise
Priorities in cloud, vendor selection,
heuristics, and best practices all indicate
that security is a leading and largely
integrated concern in advanced IT
analytics, and AIOps in particular.

Top-Down Deployment
and AIA/AIOps Evolution
6. Top-down for everything is the
winning strategy
• It is also the most pervasive.
• The executive suite (VP and above) was
more likely to be successful, and more
likely to drive, AIA strategies, deployment
and purchasing decisions.
7. Seven: Advanced analytics are showing strong
evolutionary values compared to prior years
• EMA research from early 2016 and 2018 indicate strong growth in
heuristics, data sources, integrations, stakeholder roles, and overall
versatility in terms of function and purpose.
• Although common roadblocks remain in terms of ease of use,
data management challenges, and products ‘not fully baked’

Questions
Report available at
http://bit.ly/2SGZcjt

AIOps and IT Analytics at the Crossroads: What’s Real Today and What’s Needed for Tomorrow

More Related Content

What's hot (20)

Similar to AIOps and IT Analytics at the Crossroads: What’s Real Today and What’s Needed for Tomorrow (20)

More from Enterprise Management Associates (20)

Recently uploaded (20)

AIOps and IT Analytics at the Crossroads: What’s Real Today and What’s Needed for Tomorrow