Security Encyclopedia
DDoS Attack Vector Encyclopedia
The definitive guide to every DDoS attack type - how they work, real-world impact, and how to defend against them. Built for security engineers, SOC analysts, and infrastructure teams.
OSI Reference
Attack Layer Comparison
DDoS attacks target different layers of the OSI model. Understanding which layer an attack targets is critical for selecting the right mitigation strategy.
| OSI Layer | Name | Protocol Examples | Attack Vectors |
|---|---|---|---|
| 7 | Application | HTTP, DNS, SMTP, FTP | HTTP FloodSlowlorisSlow POSTHTTP/2 Rapid ResetDNS Query FloodDNS Water TortureSSL/TLS ExhaustionWP XML-RPCAPI Abuse |
| 6 | Presentation | SSL/TLS, MIME | SSL/TLS Exhaustion |
| 5 | Session | NetBIOS, PPTP | Rarely targeted directly |
| 4 | Transport | TCP, UDP | SYN FloodSYN-ACK FloodACK FloodRST FloodFIN FloodUDP FloodTCP Fragment |
| 3 | Network | IP, ICMP, GRE | ICMP FloodSmurf AttackPing of DeathIP Null AttackGRE FloodCarpet Bombing |
| 2 | Data Link | Ethernet, ARP | MAC flooding (LAN only) |
| 1 | Physical | Cables, Radio | Not applicable to DDoS |
Historical Trends
Attack Volume Timeline
DDoS attack sizes have grown exponentially over the past two decades. What was once considered massive is now routine.
| Year | Peak Attack Size | Notable Incident | Scale |
|---|---|---|---|
| 2000 | 800 Mbps | Mafiaboy attacks Yahoo, CNN, eBay | |
| 2007 | 24 Gbps | Estonia cyberattacks | |
| 2010 | 100 Gbps | WikiLeaks-related attacks | |
| 2013 | 300 Gbps | Spamhaus DNS amplification | |
| 2014 | 400 Gbps | NTP amplification era begins | |
| 2016 | 1.2 Tbps | Mirai botnet attacks Dyn DNS | |
| 2018 | 1.7 Tbps | Memcached amplification hits GitHub | |
| 2020 | 2.3 Tbps | AWS Shield mitigates CLDAP reflection | |
| 2021 | 3.47 Tbps | Microsoft Azure UDP flood | |
| 2023 | ~3.5 Tbps / 398M rps | HTTP/2 Rapid Reset (CVE-2023-44487) | |
| 2024 | 5.6 Tbps | Record UDP flood, mitigated by Cloudflare |
Defense Planning
Mitigation Strategy Matrix
Not every defense works against every attack. This matrix maps attack vectors to effective mitigation techniques. ● = Highly effective ○ = Partially effective - = Not applicable
| Attack Vector | Rate Limiting | BGP Blackhole | Scrubbing Center | SYN Cookies | WAF Rules | Anycast | Protocol Valid. | Flowtriq |
|---|---|---|---|---|---|---|---|---|
| UDP Flood | ○ | ● | ● | - | - | ● | ○ | ● |
| ICMP Flood | ● | ○ | ● | - | - | ● | ○ | ● |
| DNS Amplification | ○ | ● | ● | - | - | ● | ● | ● |
| NTP Amplification | ○ | ● | ● | - | - | ● | ● | ● |
| Memcached Amp. | ● | ● | ● | - | - | ● | ● | ● |
| SYN Flood | ○ | ○ | ● | ● | - | ● | ● | ● |
| ACK Flood | ○ | ○ | ● | ○ | - | ● | ● | ● |
| HTTP Flood | ● | - | ○ | - | ● | ● | ○ | ● |
| Slowloris | ● | - | ○ | - | ● | ○ | ● | ● |
| HTTP/2 Rapid Reset | ● | - | ○ | - | ● | ● | ● | ● |
| DNS Water Torture | ○ | - | ○ | - | - | ● | ○ | ● |
| SSL/TLS Exhaustion | ● | - | ○ | - | ○ | ● | ○ | ● |
| Carpet Bombing | ○ | ○ | ● | - | - | ● | ○ | ● |
| Multi-Vector | ○ | ○ | ● | ○ | ○ | ● | ○ | ● |
Reference
DDoS Glossary
- Amplification Factor
- The ratio of response size to request size in a reflection/amplification attack. A factor of 50x means a 1-byte request generates a 50-byte response directed at the victim.
- Anycast
- A network addressing method where the same IP is announced from multiple locations. Incoming traffic is routed to the nearest node, distributing DDoS traffic across a global network.
- BGP Blackhole
- A routing technique that discards all traffic destined for a specific IP prefix by advertising a null route via BGP. Stops an attack but also blocks legitimate traffic.
- Botnet
- A network of compromised computers (bots/zombies) controlled by an attacker to generate distributed attack traffic. Modern botnets can include IoT devices, servers, and cloud instances.
- C2 (Command & Control)
- The infrastructure used by attackers to send instructions to a botnet. C2 channels can use IRC, HTTP, DNS, or custom protocols to coordinate attacks.
- Clean Pipe
- A DDoS mitigation service that filters malicious traffic and only forwards clean, legitimate traffic to the protected network.
- Dynamic Baseline
- An automatically calculated normal traffic profile that adapts to changing patterns over time. Flowtriq uses dynamic baselines to detect anomalies without manual threshold configuration.
- GRE (Generic Routing Encapsulation)
- A tunneling protocol that encapsulates packets inside IP. Abused in DDoS to bypass simple filtering rules or to tunnel attack traffic through network defenses.
- IOC (Indicator of Compromise)
- Observable artifacts such as IP addresses, domains, packet signatures, or behavioral patterns that indicate a security breach or ongoing attack.
- IP Spoofing
- Forging the source IP address in packets to disguise the attacker's identity or to redirect amplified responses to the victim. Essential for most reflection attacks.
- PCAP (Packet Capture)
- A file format and process for recording raw network packets. Used for forensic analysis of DDoS attacks to identify attack vectors, sources, and payload patterns.
- PPS (Packets Per Second)
- A key metric for measuring DDoS attack intensity. While bandwidth (bps) measures volume, PPS measures the processing load on network devices.
- Reflection Attack
- An attack where the attacker sends spoofed requests to third-party servers, which then send responses to the victim. When combined with amplification, this is extremely powerful.
- Scrubbing Center
- A specialized data center that filters DDoS traffic by diverting suspicious traffic for cleaning, then forwarding only legitimate traffic to the origin server.
- SYN Cookie
- A TCP defense mechanism where the server encodes state information in the SYN-ACK sequence number instead of allocating memory, preventing SYN flood resource exhaustion.
- Tbps (Terabits per second)
- Unit of measurement for network throughput. Modern record-setting DDoS attacks are measured in terabits per second - enough to saturate major internet backbone links.
FAQ
Frequently Asked Questions
A DDoS attack vector is the specific method or technique used to flood a target with malicious traffic. Each vector exploits different protocols, layers, or services - from volumetric UDP floods to application-layer HTTP attacks. Understanding attack vectors is critical for building effective defenses.
The three main categories are: (1) Volumetric attacks that saturate bandwidth (e.g., UDP floods, amplification attacks), (2) Protocol attacks that exploit weaknesses in network protocols (e.g., SYN floods, Smurf attacks), and (3) Application layer attacks that target specific services (e.g., HTTP floods, Slowloris).
UDP floods and SYN floods remain the most common DDoS attack vectors. However, amplification attacks (DNS, NTP, Memcached) have grown significantly due to their ability to generate massive traffic volumes with minimal attacker resources.
An amplification attack exploits protocols that return responses much larger than the request. The attacker spoofs the victim's IP address and sends small queries to vulnerable servers, which then send amplified responses to the victim. Memcached amplification can achieve a 51,000x amplification factor.
Flowtriq monitors packets-per-second and traffic patterns in real time at every node. It uses dynamic baselines to learn normal traffic, then detects anomalies within 1 second. It classifies the specific attack vector, captures PCAP evidence, and sends instant alerts via Slack, Discord, PagerDuty, and more.
As of 2024, the largest recorded DDoS attack peaked at 5.6 Tbps, a UDP flood targeting an East Asian ISP mitigated by Cloudflare. The previous record was a 3.47 Tbps attack mitigated by Microsoft Azure in 2021.
Carpet bombing is an advanced DDoS technique that spreads attack traffic across many destination IPs in a subnet rather than focusing on a single target. This makes detection harder because per-IP traffic may stay below thresholds, even though aggregate traffic is devastating.
Discovered in 2023 (CVE-2023-44487), HTTP/2 Rapid Reset exploits the HTTP/2 stream multiplexing feature. Attackers rapidly open and immediately cancel streams, overwhelming servers with reset processing overhead. It achieved record-breaking request rates of 398 million requests per second.
Real-Time Protection
Stop attacks before they stop you
Flowtriq detects all 35+ attack vectors in under 1 second. Classify threats automatically, capture forensic PCAP data, and get instant alerts across every channel that matters.
Start Free Trial7-day free trial · No credit card required · $9.99/node/month