2. About the Speakers
David M Davis
7 x vExpert, Author, Blogger, Speaker
Partner at ActualTech Media
Jim Shocrylas
Director of Product Management at
SIOS Technology
4. What You’ll Learn Today
Troubleshooting challenges being faced
Reasons we have troubleshooting inefficiency
Best practices for incident response
Why machine learning
Demonstration of machine learning in action
6. How Did We Get Here?
Added a virtualization layer
In an attempt to maximize datacenter resources (and
ROI), you consolidated physical hosts into VMs
(pushing up CPU and memory utilizations)
Assigned and attempted to manage virtual resources
Shared SAN LUNs between many VMs and
applications
Created the “I/O blender”
7. Key Performance Troubleshooting Challenges
Complex infrastructures
Host CPU and memory pushed to the limits
VMs moving from host to host, dynamically
VM sprawl, more and more VMs, not even sure what
they do or what their relationships are
Latency sensitive applications and latency sensitive
end users
VDI and more Tier-1 apps virtualized
8. Key Performance Troubleshooting Challenges
Limited Visibility
Current tools are designed for standalone hosts
One tool for host CPU, memory, and local disk –
another tool for the network – another tool for SAN
management – another tool for application
performance monitoring (if any)
Finger pointing
Alert storms
Trouble finding the root cause
9. Key Performance Troubleshooting Challenges
No Application Metrics
Most companies don’t have an APM
If they do have an APM it’s not virtualization-aware (yet
another tool)
Many application performance metrics are just the gut
feel of end users “it’s slower today than it was
yesterday, can you fix it?”
10. Key Performance Troubleshooting Challenges
Troubleshooting Using Trial and Error
Too much troubleshooting is done “in the dark”, i.e.”
“starting with the most likely cause (based on past
experience) and going from there…”
This is inefficient, stressful, and causes unneeded
downtime and unneeded poor application performance
Even worse – it’s easily resolved with the right tool
11. Key Performance Troubleshooting Challenges
All of this results in…
Lengthy, painful, and costly troubleshooting for your
company’s business-critical applications!
All of that equates to…
Lost revenue and wasted resources for the company
(maybe even lost customer loyalty)
12. The cost of an outage averages
$20,000 per hour
-- IDC Research
13. Best Practices for Incident Response
What you want…
Predict problems BEFORE they happen
Solve problems FASTER when they do
14. Best Practices for Incident Response
What you need….
Smarter tools
Machine analytics
15. Best Practices for Incident Response
What to look for …
Have a “Go To” tool that offers a holistic view
of the infrastructure
Don’t use threshold-based tools that
generate tons of alerts
Identify root cause – FAST
Use a tool that provides answers, not just
alerts
16. SIOS iQ™
IT Operations Analytics Platform
for Virtualized Environments
Jim Shocrylas, Director of Product Management,
SIOS Technology Corp.
17. Helping IT Meet Application Service Level Commitments
While Lowering Operations and Resource Costs
What Problems Do We Solve?
• Resolving IT issues quickly
• Maintaining resilience and availability
• Accurately planning for growth
• Accurately predicting the impact of
infrastructure or workload changes
• Lowering costs without impacting
operations
‒ Optimizing computing resources
‒ Optimizing IT’s time & effectiveness
Copyright @ 2016 SIOS Technology Corp. All rights reserved. 17
18. SIOS iQ: One Touch to Clarity
The Primary Source for IT Operations Information
• Predictive analytics without writing rules or thresholds
• Learns & predicts behavior & interactions between objects
• Provides critical intelligence with guided remediation
• Delivers immediate analysis and root cause
Groundbreaking Simplicity
• 15 minute install, minimal configuration, no agents
• Touch enabled, responsive UI
• Unifying view of infrastructure health derived through self-learning analytics
Significantly Faster Time to Issue Resolution
Automatic Identification of Issues Before They Become Problems
Copyright @ 2016 SIOS Technology Corp. All rights reserved. 18
Next-Generation IT Operations Analytics Platform
19. Today’s Monitoring Tools
Too Much Data, Not Enough Insight
• Siloed View
• Key Information Lost in Noise
• Complex, manual
• Requires Specialized Skills
• Limited Insights
• Falls Short of “Deep Root Cause”
Identification
• Too Many Different Answers
in Too Many Places
Copyright @ 2015 SIOS Technology Corp. All rights reserved. 19
• Alert Storms
• False Positives
• Averaged Data
NetworkStorageHostApplication Other
20. Automatically learns relationships and behavior patterns to
uncover hidden interactions that underlie the root cause of issues
Topological Behavior Analysis & Machine Learning
Transforms Data into a Topology Map
Measures and analyzes interrelated
objects
• Learns behavior of interrelated objects.
• Identifies anomalies
• Derives root cause and recommendations
Multi-dimensional behavior
Analysis Of Related Measurements
(Latency and IOPS for example)
Copyright @ 2016 SIOS Technology Corp. All rights reserved. 20
21. What are Next Generation IT Analytics?
Legacy
Tools
Copyright @ 2015 SIOS Technology Corp. All rights reserved. 21
SIOS iQ Next Generation
Machine Learning Analytics
Manual Configuration: Requires
configuration & continual adjustment
Self-Configuring: Requires minimal
configuration – self adjusting
Reactive: Reports current events that
IT must react to
Proactive, Preventative: Predictive
analytics identifies impending issues
Provides Data: IT left to analyze
multiple sources
Provides Intelligence: Deep analytical
understanding with recommendations
Threshold-based Rules: Limited insight
into operations causing alert storms
Self-Learning: Automatically learns
delivering only meaningful issues
22. SIOS iQ: One Touch To Clarity
VMware HA Analysis Capacity Forecast
vRealize Injection
Application Impact
Host Based Caching
PerformanceForecasting
Application Contention:
Compute, Storage, Network
Resource Contention:
Compute, Storage, Network
SnapShot Waste
VMs:
Undersized, Oversized, Idle,
Rogue
Issue
Detection
Impact Analysis
Root Cause
Identification
Solution
Recommendation
24. SIOS iQ Machine Learning-Based IT Analytics
Copyright @ 2015 SIOS Technology Corp. All rights reserved.
Protect and Optimize Business Critical Applications in VMware Environments
Unified View Across
Infrastructure
Actionable
Recommendations
One Touch Performance
Root Cause Analysis
Self Learning
SIOS PERC Dashboard™
Immediate Access to
Critical Information
Fast, Accurate Resolution
of Performance Issues
Eliminate Guesswork
Automatically Improving
& Optimizing Reported
Analysis
Analytics Derived
Information without Alert
Storms or False Positives
25. SIOS iQ: IT Operations Intelligence
in Virtual Environments
• Simple install: OVA file installs in minutes
• Analyzes Hundreds of Thousands
of Data Points Out of the Box
• One Touch Performance Root Cause Analysis
‒ Identify performance issues in one touch
‒ Optimize VMware environments for performance
‒ Meet performance requirements for business critical applications
• Host Based Caching Analysis
‒ Fast, accurate way to apply SSD storage for optimal application performance
‒ Predict potential savings and performance improvements
• VM Resource Optimization
‒ Identify idle VMs and unneeded snapshots
‒ Save money and eliminate waste
25Copyright @ 2015 SIOS Technology Corp. All rights reserved.
29. For More Information…
• Registrants are Eligible to Download a FREE 30 Day Trial of
SIOS iQ software at:
http://us.sios.com/iq/cta/free-edition/
• Email Us at: info@us.sios.com
• Follow Us on Twitter: @SIOSTech
• Call Us at 866.318.0108 (toll free US) or +1.650.645.7000
• Visit Us Online: SIOS Technology: http://us.sios.com
Hinweis der Redaktion
** slides full screen
** make us all organizers
** click SHOW MY SCREEN
** Start broadcast
** RECORD
——
[TITLE SLIDE]
Hello and Welcome to….. Webinar - Machine Learning Analytics for Immediate Resolution to the Most Challenging VIrtualization Issues
My name is David Davis and I’ll be the moderator for this event.
Today’s event is brought to you by SIOS and Actual Tech Media
Before we get started, there are just a few house keeping items I need to cover..
If you have questions during the webinar, please use the goto webinar question box to enter your question.
We’ll have having a Q&A session at the end of the event where we’ll answer those questions.
During the webinar there will be 2 points where we ask you to answer a couple of quick survey questions to ensure that we are connecting with your needs on this webinar and future webinars.
You can download a great whitepaper on this topic as well as today’s presentation in the handouts section of the goto webinar client.
We’ll be selecting one lucky attendee today to win a $300 amazon gift card that we will give away at the end of the event.
We’ve got a lot to cover so Let’s get started!!
So that you have a little background on who is presenting today, let me first tell you about myself.
My name is David Davis and I’m a VMware vExpert, VCP, CCIE, and a video training author, on the topic of virtualization, for pluralsight.com. I started my career in IT as a server and network admin, working in the datacenter. Later I was an IT manager at medium enterprise where we had a very successful server consolidation project, consolidating roughly 80% of our servers. It was then that I learned about the great power and efficiency that virtualization could provide. Since then, I’ve been writing, speaking, and creating video training around virtualization. I’ve spoke at VMware user groups and Vmworld in the US, Canada, and Europe. I’m the co-owner of actualtechmedia.com where we create technical marketing content and demand generation for companies in virtualization, storage, and cloud computing. My blog is virtualizaitonsoftware.com and you can find me on Twitter as @DavidMDavis.
I’m proud to be joined by Jim … tell us about yourself
(BIO)
Before we cover today’s agenda, I’ve got 2 quick poll questions for the audience….
How many hosts do you manage?
Do you use multiple tools to monitor your environment?
Based on what you saw today, would you like to learn more about SIOS analytics?
Key performance troubleshooting challenges facing IT managers in VMware environments
Key reasons that you may not be leveraging your infrastructure as efficiently as possible
Best practices strategy for responding to performance issues in VMware environments that aligns staff, infrastructure, and monitoring/management resources
Understand the importance of using machine learning based IT analytics as a first place to go to resolve performance issues
See a demonstration of how a machine learning based solution instantaneously identifies the root causes of performance issues, recommends solutions, and predicts the outcomes of recommendations
#FirstStopforAnswers
#FirstStopforAnswers
.
SIOS vGraph Technology
Automatically discovers complex and hidden relationships between all objects
SIOS Machine Learning Engine
Derives actionable insights
Detects anomalies
Delivers continuous analyses to derive an optimal solution
Dynamically sets thresholds and rules
Simulates effects of changes
SIOS PERC Dashboard
Presents the knowledge and advice in an easy-to-use, easy-to-access format
vGraph – discovers the complex behaviors and hidden relationships between the objects.
Machine Learning – detects the anomalies based on the patterns daily, weekly, monthly patterns…in real time.
Versus fixed or averaged computed thresholds - competitors
averages the data over 5 minute intervals
generates dynamic thresholds that are computed on weighted standard deviation
With introduction of virtualization and Software Defined Datacenter managing next generation datacenter becomes a Big Data problem. Today’s dynamic environment is getting increasingly complex with emerging technologies include Converged, Hyper-Converged, All Flash Arrays,, Host Base Caching requiring advanced approaches that understand the subtle interplay between operational real-time data in order to call out anomalous behavior which would otherwise appear as noise.
This problem requires a sophisticated analytic that is capable to extract the signal from the noise identify issues and provide the solution to a problem.
SIOS iQ takes that complex Big Data and analyzes it leveraging the principals of machine learning and patented Topological Behavior Analysis transforming that noise into the meaningful information.
1) SIOS iQ builds a multi-dimensional (internal) representation of the entire infrastructure understanding the subtle inter-relationships which taken together holistically that the less advanced approaches never discover.
2) It identifies the historical patterns of behaviors of the individual objects for individual measures such as latency, IOPs, vMem utilization etc.
SIOS iQ takes the individually learned patterns and runs it through Topological Behavior Analysis where we incorporate behavioral analysis across the multiple dimensions of the related objects determining the causal relationships to accurately and precisely delivering results which is presented to the user with a single touch.
The topology map represents the output of a small deployment taken from the SIOS iQ environment. It demonstrates the complexity, where we see two vCenters and the fanning out of VMs, DataStores from the hosts.
SIOS iQ understands the system holistically and the interrelationships among the objects in this dynamic environment as workloads are added, moved and the resulting impact.
Based on what you saw today, would you like to learn more about SIOS analytics?