Why Devops Stops
Total Page:16
File Type:pdf, Size:1020Kb
1 What is Krista? Intelligent Automation Deployment is Simple Krista is a modern conversational Intelligent Krista's Natural Language Processing supports Automation platform designed to easily leverage voice, text, and *bots to deliver automation anyone existing IT assets. Krista's unique informal understands. By utilizing existing communication approach enables business process owners to methods in conversations, you take advantage of quickly build new lookup or data entry workflows how your employees already communicate. Krista without waiting in line for expensive IT or quickly deploys to existing desktops, mobile development resources. Krista uses a unique phones, Slack, and web browsers that your programming method similar to a text conversation employees are already using. You won't need to between one or more people. By following the way train employees or maintain brittle documentation humans already communicate, Krista enables since the automation follows existing voice and anyone to build and create workflows around texting conversations similar to WhatsApp or business process constraints. The conversational Facebook Messenger. If your employees can text, workflows eliminate maintenance and upkeep they can interact with numerous systems to required from traditional record and playback support customers, consume enterprise services, automation tools. Krista's conversations are deploy IT changes, or update important KPIs. beautifully simple, with enough power, scale, and security to find any answer inside the largest enterprises. DevOps – It’s improving. DevOps Evolution Model Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Automated infrastructure Normalization Standardization Expansion Self-service delivery Many DevOps initiatives and cultures slow or stop at Stage 3 and fail to scale since organizational structures (aka people) become constraints in the Neutral Zone. Source: State of DevOps Report, 2020 4 People / Approvals in the Change Management Process Level of orthodox approvals and rates of high efficiency Orthodox change approval is based on strict 100% %High or very high inefficiency A low rate indicates higher efficiency 83% adherence to established practices: 80% 58% 60% • Changes are approved by a committee. • Approval is required from multiple levels of 40% management. 20% • Changes can be made only in predefined 9% windows. 0% Low Medium High • The person requesting the change cannot Level of orthodox approvals implement the change (separation of duties). Source: State of DevOps Report, 2020 People / Approvals in the Change Management Process Level of orthodox approvals and rates of high efficiency 100% %High or very high inefficiency A low rate indicates higher efficiency 83% Orthodox approvals make the change 80% management process less efficient. 58% 60% 40% 20% Firms with low orthodox approvals are nine 9% times more efficient than firms with high 0% Low Medium High orthodox approvals. Level of orthodox approvals Source: State of DevOps Report, 2020 People / Approvals in the Change Management Process Employee involvement and change management effectiveness 100% %High CM effectiveness 80% Firms with high employee involvement in the change management process are more than 60% five times as likely to have highly effective 39% 40% change management than firms with low employee involvement 20% 12% 7% 0% Low Medium High Level of employee involvement Source: State of DevOps Report, 2020 DevOps Evolution Model Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Automated infrastructure Normalization Standardization Expansion Self-service delivery Krista integrates your people and existing technology and tools to help scale deployments and increase collaboration amongst your people and teams. Source: State of DevOps Report, 2020 8 DevOps Nirvana MongoDB ICEFLO Chef Informix BMC RLM Coubhbase Bitbucket DB2 Cloud Application Mgr Plutora Build Master Chef ThinkSQL CouchDB GitLab Clarive Juju MySQL Ansible Automic Release Automation CircleCI Git DynamoDB Octopus Deploy Oracle EC Electric Flow AWS OpsWorks Micro Focus MarkLogic Capistrano Rational ClearCase PostgreSQL Continua CI AWS Deploy MidVision Cassandra Puppet TFS Version Control SQLite HPE Codar Codeship Puppet HBase UrbanCode Redis Run Deck Subversion SQL Server Accumulo CruiseControl SaltStack vRealize Code Stream Memcached Delphix XL Deploy Disstelli PowerShell DCS SmartFrog Kiak KV Splunk Elastic DBmaestro Actifio EC Electric-Accelerator Ehcache Jfrog Artifactory Gump Nexus Sphinx Solr Hudson CA API webSphere Tomcat Solano CI JBoss Jenkins TFS Build Bamboo Travis CI IIS GlassFish WildFly WebLogic Jetty UrbanCode Build Cucumber Selenium Team City Quick Build Shippable CA Agile Central Rocket CA Blaze Meter Jira Cloud Foundry ManisBT CA TDM FitNess CA SV Rational Test Workbench Skytap OpenVZ HP One Remedy HipChat PivotalTracker AWS Trello CoreOS CA ARD Deis SmartBear ServiceNow Mesos Azure Slack CA Flowdoc Google Docker LXC Kubernetes Sauce Labs MS Test Manager VMWare Packer BlueMix Windows Containers NuGet CA LISA Homebrew Rackspace OpenShift Triton Digital Ocean Yum Chocolatey Oracle Cloud Platform 9 Apprenda OpenStack Jelastic Helion Reality – disconnected tools with manual workflows Informix Bitbucket DB2 ThinkSQL GitLab MySQL CircleCI Git Oracle Rational ClearCase PostgreSQL Continua CI SQLite Codeship TFS Version Control Subversion SQL Server CruiseControl Disstelli EC Electric-Accelerator Gump Hudson CA API webSphere Tomcat Solano CI JBoss Jenkins TFS Build Bamboo Travis CI IIS GlassFish WildFly WebLogic Jetty UrbanCode Build Team City Quick Build Shippable Cloud Foundry Skytap AWS Deis Azure Google VMWare BlueMix Rackspace OpenShift Triton Digital Ocean Oracle Cloud Platform 10 Apprenda OpenStack Jelastic Helion Today World - Tools implemented and integration via Alt Tab Automation in siloed Need Orchestration across technology and humans MongoDB ICEFLO Chef Informix BMC RLM Coubhbase Bitbucket DB2 Cloud Application Mgr Plutora Build Master Chef ThinkSQL CouchDB GitLab Clarive Juju MySQL Ansible Automic Release Automation CircleCI Git DynamoDB Octopus Deploy Oracle EC Electric Flow AWS OpsWorks Micro Focus MarkLogic Capistrano Rational ClearCase PostgreSQL Continua CI AWS Deploy MidVision Cassandra Puppet TFS Version Control SQLite HPE Codar Codeship Puppet HBase UrbanCode Redis Run Deck Subversion SQL Server Accumulo CruiseControl SaltStack vRealize Code Stream Memcached Delphix XL Deploy Disstelli PowerShell DCS SmartFrog Kiak KV Splunk Elastic DBmaestro Actifio EC Electric-Accelerator Ehcache Jfrog Artifactory Gump Nexus Sphinx Solr Hudson CA API webSphere Tomcat Solano CI JBoss Jenkins TFS Build Bamboo Travis CI IIS GlassFish WildFly WebLogic Jetty UrbanCode Build Cucumber Selenium Team City Quick Build Shippable CA Agile Central Rocket CA Blaze Meter Jira Cloud Foundry ManisBT CA TDM FitNess CA SV Rational Test Workbench Skytap OpenVZ HP One Remedy HipChat PivotalTracker AWS Trello CoreOS CA ARD Deis SmartBear ServiceNow Mesos Azure Slack CA Flowdoc Google Docker LXC Kubernetes Sauce Labs MS Test Manager VMWare Packer BlueMix Windows Containers NuGet CA LISA Homebrew Rackspace OpenShift Triton Digital Ocean Yum Chocolatey Oracle Cloud Platform 12 Apprenda OpenStack Jelastic Helion Krista’s client interactions People Conversations / Processes Inform People Ask People Ask Systems Make Decisions Business Process Catalog Extensions / Information Read Data Analize Data Update Data Take Actions Java System A System B System C Human in the Middle Higher Levels of Automation • While we learn to trust the automation • Notifications from automation per role • Work through stability – Trust but Verify • Inclusion of humans for specific decisions • Change control needs to see history success • DevOps tooling need to be included in orchestration and strategically Automate People in manual review, • people signoff, etc. 14 Krista Orchestrating your DevOps workflow with humans Release Manager requests 1 build via Krista and review build status QA & InfoSec Release Make Go / No go Release VP makes UAT Manager decisions Manager decision Release Manager notifies QA 2 and Infosec to perform testing 2 4 QA & InfoSec have 2 hours 1 3 to perform validations in QA labs Ask Systems Make Decisions Ask People Ask People QA & InfoSec Go or No go decision is made. 4 Release manager then notifies VP of request to push to UAT 3 5 VP reviews dashboards from 5 multiple systems to make decision on publish to UAT Functional, Load, Security Systems Of Record CI/CD Build Testing QA, Release Krista watching and providing notifications Higher levels of automation with Krista 1 Release Manager requests build via Krista QA & InfoSec Release Make Go / No go Release Manager decisions Manager 2 Krista provides status from build steps to stakeholders 1 As the pipeline progresses 3 through automation, everyone Ask Systems is informed Ask Systems Inform People Inform People Make Decisions Final go / no go to UAT 4 decision based on release status 2 3 4 Functional, Load, Security Systems Of Record CI/CD Build Testing QA, Release Human in the Middle Higher Levels of Automation Value of orchestration is the key to higher • Orchestration is key to moving past levels of maturity maturity 3 • Messaging platforms get you to level 2 • Release manager did not have to and 3 understand and go into Jenkins (just make it so) • But work against you the more people you add • Humans are included in the workflow with timers and escalation, so nothing get left • Firehose of messages & spending out or stalls hours