TUT1131 - Best Practices in Deploying SUSE CaaS Platform Martin - - PowerPoint PPT Presentation
TUT1131 - Best Practices in Deploying SUSE CaaS Platform Martin Weiss Juan Utande Herrera Senior Architect Infrastructure Solutions Senior Architect Infrastructure Solutions Martin.Weiss@SUSE.com Juan.Herrera@suse.com AGENDA AGEN What
Martin Weiss Senior Architect Infrastructure Solutions Martin.Weiss@SUSE.com Juan Utande Herrera Senior Architect Infrastructure Solutions Juan.Herrera@suse.com
What What is is SUSE SUSE CaaS CaaS Platform tform 1
Requirements uirements 2 Plan Planning and Sizi and Sizing 3 Deployment B loyment Best P t Practices ctices 4 Testing ting 5 Operations rations 6
4
Physical Infrastructure: Multi-platform Servers, Switches, Storage Container Management
SUSE CaaS Platform
Storage
SUSE Enterprise Storage
Networking
SDN and NFV
Compute
Virtual Machine & Container
Multimodal Operating System
SUSE Linux Enterprise Server
Platform as a Service
SUSE Cloud Application Platform
Private Cloud / IaaS
SUSE OpenStack Cloud
Software-defined Infrastructure
SUSE Manager
Infrastructure & Lifecycle Management
Application Delivery
SUSE Global Services
Consulting Services Select Services Premium Support Services
Services
Business-critical Applications Machine Learning Business Analytics High Performance Computing Traditional IT & Applications Internet of Things
Open, Secure, Proven
Public Cloud
SUSE Cloud Service Provider Program
Where to deploy
Hardware or on your Virtualization infrastructure
Public and Private Clouds What do I need
Platform subscriptions
infrastructure nodes Who can help me
Sales Consulting:
right Hardware
solution
implementation Support options
priority support in case of issues
maintenance and proactive support to scale, upgrade, review and fix
Application Requirements (Sizing)
block, object, single or multi- writer, capacity, static or dynamic provisioning)
GPU requirements
Services / Bandwidth Security Requirements
Identity Sources Availability Requirements
$$$ BUDGET $$$ Politics, Religion, Philosophy, Processes ;-)
Kuber ernet etes es SUSE C SE CaaS P S Platform – CLUS USTER 1 R 1
Worker Worker Worker Master Admin Master Master
+
Based
number
+
Based on number of pods and resource requirements
LDAP, Salt, Velum, SQL Fault tolerance ETCD cluster
Workers as VM or physical Second cluster:
tolerance
recovery Disk Space for each Worker:
and Containers)
and changes
Review the Design
before implementation
Hardware Installation
cabling is correct
Disable everything not required (i.e. serial
ports, network boot, power saving)
Configure HW date/time VM Preparation
Preparation of Time Synchronization
Name Resolution
have different names
reverse lookup
public network
and external Velum and API targets (Cname, Load Balancer, no round robin)
Manager, RMT (limited)
same patch level on existing servers and newly installed servers
100% identical
https://github.com/Martin-Weiss/cif)
Create required Namespaces Create required Users / Groups in LDAP or Connect to Active Directory Create Roles and Role-Assignments Deploy Basic Services
Deploy Application
restore and this can not be done on the k8s side of things
cluster, re-deploy the application and restore the applications data via application functionality
Requirements uirements Plan Planning and Sizi and Sizing Deployment B loyment Best t Practices ctices Testing ting Operations rations
–Where to deploy? –On premise or public / private cloud?
–CaaS Platform Subscriptions (plus SLES for Infrastructure Service)
–For architecture and to buy the right hardware –For the initial implementation
–24/7 in case of issues
–Scale, Upgrade, Review and Fix
BENEFIT SAVINGS
cloud?
Improve operational efficiency, reduce costs, keep developers focused on development
Get to market faster, with fewer disruptions Eliminate surprises between devel-
ments—innovate faster Hardware / Virtualization Infrastructure Foster agile development and business
Integrate development and operations Manage container lifecycles
–Number of Pods –Memory, CPU –Storage requirements (file, block, s3?, single or multi-writer, capacity, static or dynamic provisioning) –specific Hardware / CPU / GPU requirements
–Network Entry points / Services / Bandwidth
–Images (source and size) –Isolation –Integration into existing Identity Sources –Certificate Authorities
–Single or multi data-center –Distance / Latency
–LDAP, Salt, Velum, SQL
–Fault tolerance –ETCD cluster
–Workers as VM or physical
–50 GB for OS (BTRFS minimum for OS) –100 GB for /var/lib/docker (BTRFS for Images and Containers) –(really depends on image sizes and image versions / image changes)
GRAYLOG PROMETHEUS GRAFANA
DATA CENTER 1 Monitoring network L2
–Implement Portus to Secure the On-Premise Registry
–Subscription Management Toolkit, SUSE Manager, RMT (limited) –Ensure staging of patches to guarantee same patch level on existing servers and newly installed servers
–Use BTRFS for the OS –Disable Firewall / AppArmor / IPv6
–Ensure that all servers are installed 100% identical –Consulting solution available (see https://github.com/Martin-Weiss/cif)
–Templates –Salt
Admin host