Diversity in DNS Performance Measures Richard Liston, Sridhar - - PowerPoint PPT Presentation

▶

diversity in dns performance measures

Diversity in DNS Performance Measures Richard Liston, Sridhar - - PowerPoint PPT Presentation

Dec 31, 2022 436 likes •831 views

Diversity in DNS Performance Measures Richard Liston, Sridhar Srinivasan and Ellen Zegura liston,sridhar,ewz@cc.gatech.edu Georgia Institute of Technology Diversity in DNSPerformance Measures p.1/22 Roadmap Background: DNS Problem:

slide-1

SLIDE 1

Diversity in DNS Performance Measures

Richard Liston, Sridhar Srinivasan and Ellen Zegura

liston,sridhar,ewz@cc.gatech.edu

Georgia Institute of Technology

Diversity in DNSPerformance Measures – p.1/22

slide-2

SLIDE 2

Roadmap

Background: DNS Problem: Measuring DNS Our method: requirements and solution Results Conclusions

Diversity in DNSPerformance Measures – p.2/22

slide-3

SLIDE 3

DNS: Domain Name System

Solution to the problem of global addressing Distributed database over hierarchical namespace Allows for local administration of Namespace (via delegation of authority over zones) Caching policies (via TTL) Details hidden from users (gethostbyname()) Supported by: root servers: .gov, .edu, .fr, ... gTLD servers: .com, .org, ...

ther servers: lower levels in the namespace

hierarchy

Diversity in DNSPerformance Measures – p.3/22

slide-4

SLIDE 4

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-5

SLIDE 5

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-6

SLIDE 6

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-7

SLIDE 7

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-8

SLIDE 8

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-9

SLIDE 9

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-10

SLIDE 10

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-11

SLIDE 11

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-12

SLIDE 12

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-13

SLIDE 13

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-14

SLIDE 14

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-15

SLIDE 15

What Happens in the Wild...

A.ROOT-SERVERS.NET B.ROOT-SERVERS.NET M.ROOT-SERVERS.NET A.GTLD-SERVERS.NET B.GTLD-SERVERS.NET M.GTLD-SERVERS.NET gatech.edu zeon.cc.gatech.edu gaia.cc.gatech.edu itchy.mindspring.net cnn.com CLIENT LOCAL DNS SERVER

Diversity in DNSPerformance Measures – p.4/22

slide-16

SLIDE 16

Resolution Tree: Critical Path

Query from client Query to gTLD Referral from gTLD Query to NS1 Query to NS2 Timeout/re-xmit to NS1 Response from NS1 Response from NS2 Response to client

Diversity in DNSPerformance Measures – p.5/22

slide-17

SLIDE 17

Resolution Tree: Critical Path

Query from client Query to gTLD Referral from gTLD Query to NS1 Query to NS2 Timeout/re-xmit to NS1 Response from NS1 Response from NS2 Response to client

Diversity in DNSPerformance Measures – p.5/22

slide-18

SLIDE 18

A B C M Client 1 ... A B C ... M GTLD Servers: Client 2 Root Servers: Domain Servers: 3 1 2 2 1 example.com example.net example.org

Diversity in DNSPerformance Measures – p.6/22

slide-19

SLIDE 19

Diversity in DNSPerformance Measures – p.7/22

slide-20

SLIDE 20

Measuring DNS

Primary question in this study: To what extent do DNS performance measures for non-cached names vary across Internet locations? Measures: Overall response time Root and gTLD server response time Aliasing via CNAME TTL (Time-To-Live in seconds)

Diversity in DNSPerformance Measures – p.8/22

slide-21

SLIDE 21

Measurement Method

Criteria: Capture fine-grained information about DNS system operation Compare data collected at different sites Easily collect data at multiple locations

Diversity in DNSPerformance Measures – p.9/22

slide-22

SLIDE 22

Measurement Method

Criteria: Capture fine-grained information about DNS system operation Compare data collected at different sites Easily collect data at multiple locations Solution: Modify BIND to log relevant info: tx/rx packets, packet type, timeouts, timestamp Package modified BIND w/non-root config files and driving script w/15K domain names Use dig to query the modified server Domain names are diverse, non-cached Post-process logs

Diversity in DNSPerformance Measures – p.9/22

slide-23

SLIDE 23

Data Collection

NIMI nodes, colleagues and Linux users; 75 sites in 21 countries and territories (still need China and India!) Different days of week and times of day Variety of connection technologies: DSL, PPP, cable modem, GigE, etc. January 2002 and late March/early April 2002 Single run: 4 to 6 hours Very lightweight

Diversity in DNSPerformance Measures – p.10/22

slide-24

SLIDE 24

Validating Workload Set

Completion and success rates Response codes: 0: No error 2: Remote server failure 3: Name does not exist Completed (0 or 3): 96.4% to 98.1% Successful (0): 92.7% to 94.7%

Diversity in DNSPerformance Measures – p.11/22

slide-25

SLIDE 25

Mean Response Time, com- pleted (MRTc)

20 40 60 1.0 1.2 1.4 1.6 1.8 2.0 2.2

Sites Mean response time of completed lookups (sec)

Range from 0.95 seconds to 2.31 seconds!

Diversity in DNSPerformance Measures – p.12/22

slide-26

SLIDE 26

Measuring Major Factors

Connectivity MINc: Minimum response time for completed lookups Minimum round trip time to closest name server queried

Diversity in DNSPerformance Measures – p.13/22

slide-27

SLIDE 27

Measuring Major Factors

Connectivity MINc: Minimum response time for completed lookups Minimum round trip time to closest name server queried Loss Rate Total retries along all critical paths

Diversity in DNSPerformance Measures – p.13/22

slide-28

SLIDE 28

Measuring Major Factors

Connectivity MINc: Minimum response time for completed lookups Minimum round trip time to closest name server queried Loss Rate Total retries along all critical paths Root/GTLD server performance MRTr/MRTg: mean response time to root/gTLD servers

Diversity in DNSPerformance Measures – p.13/22

slide-29

SLIDE 29

Measuring Major Factors

Connectivity MINc: Minimum response time for completed lookups Minimum round trip time to closest name server queried Loss Rate Total retries along all critical paths Root/GTLD server performance MRTr/MRTg: mean response time to root/gTLD servers Location relative to other nameservers MRTl: mean response time to a fixed set of last nameservers contacted

Diversity in DNSPerformance Measures – p.13/22

slide-30

SLIDE 30

Correlations

Factor Correlation w/MRTc Connectivity (MINc) 0.62 Loss Rate (# retries) 0.50 Root Server Performance (MRTr) 0.86 GTLD Server Performance (MRTg) 0.94 Location (MRTl) 0.90

Diversity in DNSPerformance Measures – p.14/22

slide-31

SLIDE 31

Impact of Root/GTLD/Other

Sites Percentage of access time

20 40 60 80 100

ther

gtld root

Diversity in DNSPerformance Measures – p.15/22

slide-32

SLIDE 32

Some Implications

Roots: poor performance barely noticable GTLDs: poor performance more noticable ( 20% - 30% of time spent in querying gTLD servers); better server placement could help Other servers: most potential for performance enhancement; do CDNs want to provide DNS-only services??? A service differentiator for ISPs may involve performance from gTLD servers ISPs may consider choosing different routes to gTLD servers

Diversity in DNSPerformance Measures – p.16/22

slide-33

SLIDE 33

More on root/gTLD Interaction

BIND employs a server selection algorithm Maintains history of NS response times Ages performance history to sample all servers Converges on one or more servers "Favored" servers: >= 10% of queries

Diversity in DNSPerformance Measures – p.17/22

slide-34

SLIDE 34

Root Servers

A B C D E F G H I J K L M

Sites Root servers chosen for > 10% of queries from each site

Diversity in DNSPerformance Measures – p.18/22

slide-35

SLIDE 35

gTLD Servers

A B C D E F G H I J K L M

Sites gTLD servers chosen for > 10% of queries from each site

Diversity in DNSPerformance Measures – p.19/22

slide-36

SLIDE 36

Other Results: CNAME/TTL

Measure Observations CNAME

Fairly consistent use across sites 26% of our names were aliases 6% resolved to more than one different CNAME 1% (four) resolved to more than ten different CNAMEs

TTL

Distribution the same across all sites Most popular TTLs: one day, one hour, two hours, 12 hours, two days

Diversity in DNSPerformance Measures – p.20/22

slide-37

SLIDE 37

Conclusions

Most widespread measurement of DNS performance to date Measures controlled by admins show low variation: CNAMEs, TTLs Measures subject to network location vary widely Mean response time for completed lookups Mean response time from root/gTLD servers Favored root/gTLD servers Potential for performance enhancement: gTLD/other servers Demonstrated need for multiple measurements

Diversity in DNSPerformance Measures – p.21/22

slide-38

SLIDE 38

Still collecting data! http://www.cc.gatech.edu/ liston/dnsperf.html Questions?

Diversity in DNSPerformance Measures – p.22/22