← Plain Library

Security monitoring · Plainly explained

SIEM Security Information & Event Management

Bring security events from across your infrastructure into one place, connect the dots, detect suspicious activity, and alert your security team.

SIEM collects logs from different systems, analyzes them, correlates related events, and generates security alerts when suspicious activity is detected.

Explore how SIEM works
Firewall
EDR
Windows
Linux
Cloud
Applications
Network
LIVE ANALYSISSIEMCollect · Connect · Detect

01 / WHY SIEM?

Every tool sees a fragment. SIEM sees the story.

A modern IT environment contains many security tools, servers, apps, identities, and cloud services. Each one records only its own part of what happened. Those records are called logs.

FW

Firewall

EDR

EDR / Antivirus

ID

IDS / IPS

VPN

VPN

AD

Active Directory

WIN

Windows Servers

LIN

Linux Servers

NET

Routers & Switches

CLD

Cloud Platforms

SaaS

SaaS Applications

DB

Databases

IAM

Identity & Access

APP

Business Applications

01VPN login02AD authentication03EDR activity04Firewall traffic
POSSIBLE THREATAccount compromiseFour related events · one security story
THE CENTRAL IDEASIEM connects the dots.

It can collect useful logs from security products and from ordinary IT systems, applications, databases, cloud platforms, and devices.

02 / ROLE OF SIEM

A secondary line of defence—watching while events unfold.

SIEM is primarily a monitoring and detection technology. It observes what is happening, identifies suspicious activity, and tells the security team what deserves attention.

01 · PREVENTION / PROTECTIONFirewall · EDR · Antivirus · IPS · MFA

Controls designed to block, limit, or resist threats.

02 · DETECTION / VISIBILITYSIEM

Observes activity and detects suspicious patterns as events occur.

03 · INVESTIGATIONSOC

Analysts decide what happened and how serious it is.

04 · RESPONSESOAR / Security controls

People and integrated tools contain and remediate the threat.

CORE CHARACTERISTIC

SIEM is largely passive in nature.

It captures events, collects logs, correlates activity, detects suspicious patterns, generates alerts, and provides visibility. It does not inherently prevent every attack.

  • Can detect activity during an attack
  • Does not need to wait until an attack is over
  • Focuses on monitoring, detection, and alerting
  • Can integrate with SOAR, EDR, firewalls, IAM, and other response systems

03 / WHAT ARE LOGS?

A log is a record of something that happened.

Logs are the raw material of a SIEM. A useful event usually answers several simple questions: who acted, what happened, when and where it happened, and whether the action succeeded.

event.log● ● ●
User:        john
Source IP:   10.10.10.25
Destination: Server01
Action:      Login
Result:      Failed
Time:        10:30:15
WHO

User

john
WHAT

Action

Login
WHEN

Time

10:30:15
WHERE

IP / Device

10.10.10.25
RESULT

Outcome

Failed

04 / LOG COLLECTION

Before SIEM can analyze a log, the log must reach it.

Different technologies use different collection methods. Some send events continuously, while others let a collector retrieve them on a schedule.

SYS01

Syslog

A common way for network and security devices to send event messages.

Firewall
Syslog
Collector
SIEM
Collection notes

514 uses UDP or TCP. 6514 commonly uses TLS encryption.

Firewalls · routers · switches · IDS/IPS · Linux · appliances
WEF02

Windows Event Forwarding

Windows Event Logs can be forwarded with WEF, commonly over WinRM.

Windows
WEF / WinRM
Collector
SIEM
Collection notes

Authentication can use Windows credentials or other supported mechanisms.

Windows clients · Windows servers · domain controllers
AGT03

Agent-based

Installed software reads local logs and forwards them to a collector or SIEM.

Endpoint
Agent
Collector
SIEM
Collection notes

Agents may filter, buffer, encrypt, or preprocess events before sending them.

Servers · endpoints · applications · custom log files
API04

REST API

A collector retrieves events from an HTTP-based interface exposed by a service.

Cloud / SaaS
REST API
Collector
SIEM
Collection notes

API credentials, tokens, certificates, and rate limits may control access.

Cloud · SaaS · identity platforms · security platforms
DB05

Database

A collector reads event records stored by an application in a database.

App database
ODBC / JDBC
Collector
SIEM
Collection notes

The connection should use a narrowly scoped account and protected transport.

Business apps · audit databases · legacy systems

05 / LOG FORMATS

Transport is the journey. Format is the shape of the data.

These ideas are often mixed together. Transport describes how the event travels. Format describes how its data is organized.

TRANSPORTHow the log travels

Syslog connection · WEF · agent · REST API · database connection

FORMATWhat the log looks like

Syslog message · CEF · LEEF · JSON · CSV · Windows Event XML

Syslog

Common text-based event format used by systems and network devices.

CEF

Common Event Format: a structured security-event format.

LEEF

Log Event Extended Format: another structured security-event format.

JSON

Key-and-value format used by modern apps, cloud services, and APIs.

CSV

Comma-Separated Values: data arranged into rows and columns.

Windows Event / XML

Structured Windows events with fields such as Event ID, provider, user, process, and time.

event.jsonJSON
{
  "user": "john",
  "source_ip": "10.10.10.25",
  "action": "login",
  "status": "failed"
}
event.csvCSV
Time,User,SourceIP,Action,Status
10:30:15,john,10.10.10.25,login,failed

WINDOWS EVENT FIELDS

Event IDProviderComputerUserTimeProcessCommand

06 / PUSH VS PULL

Who starts the conversation?

That one question separates the two collection models.

PUSHSOURCE INITIATES
DEVICECOLLECTOR

The source sends logs to a destination.

  • Destination IP or hostname
  • Destination port and protocol
  • Security settings where applicable
Firewall → Port 514 → Collector
PULLCOLLECTOR INITIATES
COLLECTORDEVICE / CLOUD

The collector connects to the source and asks for logs.

  • Source IP, hostname, and port
  • User ID and password where used
  • API credentials, tokens, or certificates
Collector → REST API → Cloud logs
Push = Source sends the data.Pull = Collector asks for the data.

07 / COMPLETE ARCHITECTURE

From scattered events to a searchable security story.

Follow the main path down. Select any component to open a plain-language explanation. The archive branch keeps events for later investigation, compliance, auditing, and historical analysis.

01Event Sources+

Systems that create security or operational events.

Firewall · EDR · Windows · Linux · cloud · apps · databases · network devices
02Log Collector+

Receives pushed logs or retrieves them from sources.

Syslog · WEF · agents · REST APIs · database connections
03Decoder / Parser+

Turns raw text into meaningful fields the SIEM can understand.

User · action · result · source IP · time
04Metadata & Normalization+

Extracts useful details and maps different field names to common concepts.

src_ip · IpAddress · sourceIPAddress → Source IP
05Concentrator / Indexing+

Organizes events so searches across large volumes remain fast.

Find every failed login by John in the last 24 hours
06Event Stream Analyser+

Checks events as they arrive for rules, patterns, thresholds, and correlations.

Real-time analysis · rule matching · pattern matching · threat detection
07Security Analytics+

Gives analysts the views and tools used to investigate.

Dashboards · alerts · search · reports · threat hunting · compliance
08SOC+

The people and processes that triage alerts, investigate evidence, and coordinate response.

Triage · investigation · decision · escalation · response

ARCHIVE BRANCH

ArchiverRetains event data for later use.
HOTRecent · frequently accessed
WARMOlder · less frequently accessed
COLDHistorical · long-term retention
RAW EVENTUser john failed login from 10.10.10.25
PARSED + NORMALIZED
User
john
Action
Login
Result
Failed
Source IP
10.10.10.25

WHY NORMALIZATION MATTERS

Different names. One common concept.

Mapping equivalent fields makes correlation across unrelated products much easier.

Firewall src_ipWindows IpAddressCloud sourceIPAddressSource IP

08 / RULES & CORRELATION

Where SIEM becomes smart.

A rule defines a condition that should trigger an alert. Correlation connects related events to reveal a larger security story.

RULE CONDITIONMore than 10 failed logins from the same IP within 5 minutesALERT
01VPN login from unusual location02Successful authentication03Privileged account used04Critical server access05Large data transfer
CORRELATION
POSSIBLE INCIDENTAccount compromise / data exfiltration
One event may not be suspicious. Several related events together can tell a very different story.
01

Threshold rules

More than 10 failed logins in 5 minutes.

02

Sequence rules

Failed login → success → privilege escalation → sensitive file access.

03

Correlation rules

VPN + Active Directory + EDR + firewall activity.

04

Pattern-based rules

PowerShell + encoded command + suspicious network connection.

05

Time-based rules

50 authentication failures within 2 minutes.

09 / FROM LOG TO ALERT

One event. Fourteen stages. One informed decision.

The complete path shows how raw activity becomes something an analyst can investigate.

  1. 01Event occurs
  2. 02Source generates log
  3. 03Log transport
  4. 04Log collector
  5. 05Decoder / parser
  6. 06Metadata extraction
  7. 07Normalization
  8. 08Indexing / storage
  9. 09Correlation & detection
  10. 10Rule match
  11. 11Alert
  12. 12SOC analyst
  13. 13Investigation
  14. 14Response

10 / UNDERSTANDING ALERTS

An alert is a question—not proof of an attack.

A SIEM says, “This activity matches something worth checking.” A security analyst must investigate the context and decide what actually happened.

10 FAILED LOGINSObserved activity
SIEM ALERTRule matched
INVESTIGATIONAnalyst checks evidence
Brute-force attackUser forgot passwordApplication issueAutomated service
FALSE POSITIVE

The SIEM alerts, but the activity is legitimate.

AlertInvestigationNormal activity
FALSE NEGATIVE

A real attack occurs, but the SIEM does not detect it.

AttackNo alert

Good SIEM engineering tries to reduce both.

SIEM ALERT LIFECYCLE

Detection begins the work. It does not finish it.

  1. 01Event
  2. 02Detection rule
  3. 03SIEM alert
  4. 04Triage
  5. 05Investigation
  6. 06True / false positive
  7. 07Incident response
  8. 08Containment
  9. 09Remediation
  10. 10Recovery
  11. 11Lessons learned

11 / SIEM & SOC

SIEM is the technology. SOC is the team around it.

A SOC combines people, processes, and technologies to monitor and respond to security incidents. Analysts triage alerts, collect evidence, assess impact, and coordinate the next action.

LogsSIEMAlertsSOC analystInvestigationResponse

12 / SIEM VS SECURITY TOOLS

SIEM brings their information together—it does not replace them.

TechnologyPrimary purpose
FirewallControls network traffic
AntivirusDetects and blocks malware
EDRMonitors and responds to endpoint activity
IDS / IPSDetects or prevents suspicious network activity
IAMManages identities and access
DLPHelps protect sensitive data
SIEMCollects, correlates, analyzes, and alerts
SOARAutomates and orchestrates security response

13 / SIEM + SOAR

Detection meets automated response.

SIEM focuses on monitoring and detection. SOAR coordinates tools and automates approved response steps.

SIEMDetects malicious IP
SOARReceives alert
APICalls firewall
BLOCKIP blocked
SOCTeam notified

Modern SIEM platforms can integrate with SOAR, EDR, firewalls, IAM, and other systems to enable automated response. Automation should still follow carefully reviewed rules because a mistaken action can disrupt legitimate users.

14 / WHAT MAKES SIEM EFFECTIVE?

Technology is only one part of the system.

A SIEM is only as effective as the data, detection rules, and processes around it.

  • 01Good log coverage
  • 02High-quality logs
  • 03Correct parsing
  • 04Proper normalization
  • 05Useful correlation rules
  • 06Low false-positive rate
  • 07Appropriate log retention
  • 08Useful dashboards
  • 09Well-tuned alerts
  • 10Skilled SOC analysts

15 / WHAT SIEM DOES NOT DO

SIEM is not a magic security product.

It cannot guarantee that every attack will be detected.

Detection can fail when…

  • A device does not generate the required log
  • Logs are not being sent
  • Logs are incorrectly parsed
  • Important fields are missing
  • Detection rules are poorly configured
  • Alerts are ignored
SIEM visibility depends on log coverage, log quality, detection engineering, and analyst investigation.

16 / FINAL SUMMARY

SIEM in 7 simple steps.

01

Collect

Get logs from different systems.

02

Transport

Bring logs in through Syslog, WEF, agents, APIs, databases, and more.

03

Parse

Understand the structure and fields inside each log.

04

Normalize

Map different log structures into common security concepts.

05

Index & store

Organize events for fast search and suitable retention.

06

Correlate & detect

Connect related events and apply detection rules.

07

Alert & investigate

Tell the SOC what needs attention so analysts can respond.

THE SHORT VERSION

Think of SIEM as the security control room of your IT environment: many systems send it information, SIEM connects that information, identifies suspicious activity, and brings the important events to the attention of the security team.

Return to the top