Prof Messer CompTIA Sec+ SY0-701 3.4

high availability

redundancy doesnt always mean available

always on, always available

almost always means higher costs

server clustering

combine multiple servers to appear and operate as 1

- users only see 1 device

- servers are aware of each other

easily scale

usually configured in the OS of servers

servers usually have a shared storage they are connected to

load balancing

distributes load over multiple servers

servers are usually unaware of each other

can add or remove devices

site resiliency

recovery site is prepped

- data synched

ready for disaster events

hot site

exact replica of everything

- buy 2 of everything every time

apps and software, etc all the same

ready to be used

cold site

no hardware; empty building

bring data and people with you

warm site

somewhere between cold and hot site

- bring some things, not all

geographic dispersion

sites should be far apart

can be a logistical challenge

platform diversity

every OS has security issues

many vulnerability are specific to a single OS

use diff platforms/OS's

- spread the risk around

multi-cloud systems

many cloud providers

plan for outages

if issue happens, your data is spread out

continuity of operations planning (COOP)

alternative way of staying up when tech goes down

- manual transactions

- paper receipts

- phone calls for approvals, etc

must be documented and tested beforehand

capacity planning

match supply with demand

too much demand

- apps shut down, outages

too much supply

- wasted money

balanced approach

people (capacity planning)

some services need people

too few

- recruit new staff and train

- can be time consuming

too many

- redeploy

- downsize

tech scalability (capacity planning)

pick tech that can scale

web services

- distribute load across multiple services (load balancers)

database services

- cluster servers

- split database across multiple servers

cloud services

- on demand

- seemingly unlimited resources (if you $$)

infrastructure (capacity planning)

underlying framework

- app servers, network storage, etc

physical devices

- purchase, configure, install

cloud based devices

- easier to deploy

- useful for capacity changes

recovery testing

testing yourselves before an actual disaster

evaluate response and make changes

tabletop exercise

going through the steps of the disaster drill but not actually doing it

- actually doing it can be expensive

fail over test

test used to see if redundant configs would switch over in the event of a failure

simulation testing

testing the systems with simulated attacks

- phishing, data breaches, etc

parallel processing

split a process through multiple CPUs

can be a single computer with multiple CPU cores or multiple comps

improved performance and recovery

backup factors

amount of data

type of backup

storage location

backup media

day of the week

backup and recovery software

onsite vs offsite backups

on site backups

- data immediately available

- generally less $$

- no internet link needed

off site backup

- transfer data over internet or WAN link

- data is available after disaster

- restoration can be done anywhere

orgs usually have both

Backup frequency

how often data is backed up

may be diff between systems

diff sets of backups

- daily, weekly, monthly

lots of planning needed

encrypting backups

backups are often encrypted

recovery keys needed to restore data

very useful for cloud backups and storage

snapshot backups

popular on VMs and cloud

an instant backup of an entire system

-save current configs and data

backups contain the changes between snapshots

- take one every day

replication backup

ongoing almost real time backup

- data is synched in multiple locations

- changes are updated to copies

always a copy available somewhere

journaling

used when stored data is corrupted

- recovery is complicated (remove corrupted files, restore)

before writing to storage, write to a journal

- write to storage after

if power runs out while journal is writing to storage, system can use the journal to update the database

if power goes out while data is going to journal then the data is lost

- database has no corruption

UPS

Uninterruptible Power Supply

-short term backup power

- blackouts, brownouts, surges

UPS types

- offline/standby UPS

- line interactive UPS

- on line/ double conversion UPS

features

- auto shutdown, phone line suppression, etc

offline/standby UPS

Only goes online if you lose the main source power.

line interactive UPS

can slowly increase the amount of voltage if main power line has a drop

on line/ double conversion UPS

always online & providing power

generators

long term power backup

- fuel needed

may take a couple mins to ramp up after outage

- UPS can cover