VL 4 Flashcards
AWS - Resource Distribution
Scalability, fault tolerance in availability because of many regions
AWS - Availability Zones
Two zones have no common points of failure => servers in two zones gain infrastructural redundancy
Communication in one zone is free, between zones not
AWS - SLA
Service Level Agreement
Region level: region is unavailable, when all of your running instances or tasks, deployed in two or more AZs, concurrently have no external connectivity
Instance- level: instance is not available, when your Single EC2 instance has no external connectivity
Xen Hypervisor
3 levels of virtualization: bare metal, hosted, os level
Xen: bare metal Hypervisor
Nitro Hypervisor
Services are now part of the hardware
Amazon Machine Image (AMI)
Copy of server with OS and pre installed software
AMIs are stored in S3
AWS Storages
Amazon Elastic block storage: physical disks accessible to the network
Amazon EC2 Instance Storage: physical disks connected to the server on which VM will be placed
Amazon Elastic File System: Network attached system
Amazon simple storage service: place blob of data
RAID
Increase bandwidth, fault tolerance
(Create copies)
Local storages
RAID
FLASH: non-volatile memory
Single and Multi level Cells
Data center storage
Only flash or SSD
Provisioning of storage
Direct attached storage
Storage area network: independent from LAN
Network Attached Storage
SAN pros
Flexible distribution of devices between clients
Easy replacement of faulty servers with new servers booting the same unit
Easier disaster protection
SAN cons
Shared network bandwidth
Shared performance of storage devices
Security concerns due to transfer data through network
Network attached storage (NAS)
File server attached to the network
Frequently using internally a RAID to get redundancy
Storage virtualization
Block virtualization or file virtualization
Usage:
Flexible mapping, thin provisioning, disk expansion and shrinking, non-disruptive data migration, improved utilization