Networking/Computing Tips/Tricks

Have you noticed a lot of QUIC protocol in your packet captures?  I certainly have, and we had better talk about what this is.

What is QUIC?

QUIC is a new protocol created by the fine folks at Google.  It stands for Quick UDP Internet Connections.  It is an experimental protocol that was introduced back in 2012 on the Chrome and Chromium platforms, and is getting broader use by the day as users continue to flee from Microsoft's Internet Explorer and Edge.  Google calls it a transport layer protocol (which would be Layer 4) but it actually runs over UDP which is at Layer 4 - so we see it as a new Layer between 4 and 5.

The target of QUIC, although Google soft shoes this point, is to essentially replace TCP.  TCP has been around a long time, and while it has received many updates to improve function, speed, and scalability, due TCP's dependency on round-trip exchanges and acknowlegements (ACK's), many believe it is becomming outdated based on today's higher speed demands of the Internet.  By high speed demands, what we mean is that in the original web pages of the Internet, life was simple, it took one TCP session to send the page.  Today, it can take dozens of TCP connections to load all the elements on a single web page.  These multiple connections take time and have to be processed in order.  TCP has no ability to multiplex these connections.

In fact, one of the problems that can occur is that let's say the first of several dozen TCP connections is slow or ties up the resources of the receiver.  Other processes won't even start!  So that first process essentially blocks other processes from even starting.  Google calls this 'head of the line blocking'.

One attempt to improve this multiplexing was a protocol called SPDY (speedy).  But SPDY did not fix all the problems.  You can read more on SPDY here.

Let's get back to QUIC.  QUIC supports a set of multiplexed connections between two endpoints over UDP, and was designed to provide security protection equivalent to TLS/SSL.  So along with reduced connection and transport latency, and bandwidth estimation in each direction to avoid congestion, QUIC's achieves its main goal to improve perceived performance of connection-oriented web applications that are currently using TCP.

Surprisingly there are approximately 30 internet draft specifications dealine with QUIC.  You can find those drafts here.

Google's Own Explanation of what QUIC is

You can also view Google's own explanation about QUIC in the following video:

 

 Examining QUIC on Your Computer

One of the cool things that Google does for us is it allows us to view QUIC performance within Chrome.  So if you have Chrome installed, open a new tab and enter the following:

 chrome://net-internals/#quic

You will something like this:

QUIC1

 You can see the number of events, the QUIC version that you are running, and Connection ID's.  This screen will update as connections come and go.  

If you click on one of your connection ID's, and then on the next page, click on the Source Type field, you will get something that looks like this:

QUIC2

To the right, you will see the actual packet process that QUIC is using as all the protocol processes are dissected (well, without the data).  This is a live screen, and it will continue to update if there is data going in the session.

Wireshark does have the ability to dissect QUIC processes as well (ofcourse it cannot decrypt the data), and we will discuss this in another article.

Scanning through the packet information you will see interesting information like packet numbers, missing packet information, acknowleged packet information and much more.

We hope this article has helped you to understand what QUIC is at an introductory level.  Look for more details to come by simply clicking on the QUIC tag below.

Comments?  Questions?

 

 

 

Comments powered by CComment

Subscribe to our Newsletter!

Our Tag Cloud

4G Networks 6LoWLAN 6LoWPAN 802.11 802.11ah 802.11ax 802.11ay 802.11az Ad-Hoc Addressing Airlines Analysis Ansible Apple Architecture ARP Assessment ATM AToM Automation Baseline BGP Billing Bloom's Taxonomy Bluehost BPF Briefings Cable Capture Filter CellStream Cellular Central Office Cheat Sheet Chrome Cisco Click Model Cloud CMD Company Policy Computer Consulting Data Center Data Networking Decryption Dependencies DHCPv6 dig Display Filter DNS Documentation Earth Earthquakes Ethernet Ethics Etiquette Evaluation Filter Five Monkey Rule G-MPLS Gauge GNS3 Google Hands-On Hiring History Home Network HTTPS ICMP ICMPv6 IEEE 802.11p IEEE 802.15.4 India Internet IoT IPv4 IPv6 IRINN IS-IS L2VPN L3VPN LDP LifeNet Linux LLN LoL M-BGP MAC Macro Management Microsoft Milky Way mininet Monitoring MPLS mtr Multicast Murphy Netcat NetMon netsh Networking nmap nslookup Observations OLPC Online School OpenFlow OSPF OSPFv2 OSPFv3 OSX OTT Personnel Policy POTS POTS to Pipes PPP Profile Project Management PW3E QoS QUIC Railroad Remote Desktop Requirements Review RIP Routig Routing RPL RSVP Rural Scanning SDN Security Service Provider Small Business SONET Spam Speed SS7 SSL Status Storms Subnetting Support T-Shark TCP TCP/IP Telco Telecom 101 Telecommunications Telephone Testing TLS Tools Traceroute Traffic Engineering Training TRANSUM Travel Tunnel Ubuntu Utility Video Virtualization VoIP VRF Wi-Fi Windows Wireless Wireless 5G Wireshark WLAN ZigBee

Our Twitter Feed

SiteLock