A High Level look at RTB1. Browsers visit Publishers and create impressions.2. Publishers sell impressions via Exchanges.3. Exchanges serve as auction houses for the impressions4. On behalf of the marketer,m6d bids the impressions via the auction house. If m6d wins, we display our ad to the browser.
Performance and Data• Billions and billions of bid requests a day • A single request can result in multiple Cassandra Operations! • One cluster is just under 10TB and growing• Low latency requirement below 120 ms typical• Limited data available tom6dvia the exchange
Segment DataSegments are how we assign product or serviceaffinity to a group of users. User’s we consider to belike minded with respect to a given brand will beplaced in the same segment.Segment Data is just one component of ouroverarching data model.Segments help to reduce the number of calculationswe do in real time.
Old Approach for Segment Data Application Nodes (Tomcat + MySQL ) Limitations •Periodically updated.MySQL Data Push Event Logs •Only subsection of the data. •Cluster performance is effected during a data push. Aggregation Hadoop
Cassandra Approach for Segment DataApplication Nodes Better! (Tomcat + Less • Updating in real time now MySQL Usage) possible • Distributed not duplicated • Lesscomplexity to manage • Storing more information • We can now bid on users Cassandra sooner!
During waking hours: Dr. Realtime• User traffic is at peak• Applications need low latency operations• High volume of read and write operations• Desire high cache hit rate to limit disk IO• Dr. Realtime conducts experiments on optimization
Experiment: Active Set, VFS, cache size tuning• Cluster optimization is a topic that must be revisited periodically• User base and requests are perpetually growing• Amount of physical data stored grows• New features typically result in new data and more requests• How to tune your environment is application and hardware dependent
Physical data directory• sstable holds data• Index holds offsets to avoid disk seeks• Bloom filter probabilistic lookup system – (also a stat table)
When RAM > Data Size• If you can afford to keep your data set in RAM:• It is fast from VFS cache• Thats it. Your optimized.• However you do not usually need this much ram
When RAM < Data Size• The OS will cache the most active portions of disk• The write/compact model causes the cache to churn• User requests causes the cache to churn
Understanding Active set with a hypothetical exampleWebmail service (Coldmail): • I have an account for 10 years, I never log in more than twice a month • I have 1,000,000 items in my inbox • Not in the active setSocial networking (chirper): • I am logged in every day • Commonly read get updates from my friends • In the active set
$60,000 QuestionHow do you determine what theactive set of your application anduser base is?
Setup instruments for testing
Turn on a cache• JMX allows you to tune only a single node for side by side comparisons• Set the size very large for key cache (be more careful with row cache)
Analysis • 8:30 hit rate 91% 1.2 mil • 10:30 hit rate ~93% 1.7 mil • Past 1.2 million entry cache might be better spent elsewhere
Active set conclusions• Determine sweet spot for hit rate and cache size• Do not try to cache long tail of requests• When all other things equal dedicate more cache to most read column family• Use row cache only if rows are a predictable size• Large row caches can not be saved so cold on restart
read_repair_chance – Cassandras version of an ethical dilemma• Read Repair generates additional reads across the cluster for each user read• Read Repair Chance controls the probability of Read Repair occurring.• If data is write-once or write-rarely Read Repair may be unnecessary – data read ratio much larger then write ratio – data that does not need strict consistency• 1.0 Hinted handoff now does not need to wait on the failure detector. Read Repair Chance default has been set to 10% from 100%. – Cassandra-2045 TX ntelford and co!
Analysis for RRC test subjects Candidate: Many reads few writes Inside story: This data used to take 2 days. A few ms... Come on man! Candidate ?: Many writes Inside story: This is used for frequency capping, higher % justified
Experiment: Test the limits of NoSQL science with YCSBYCSB is a distributed load generatorthat comes in handy!• Before our upgrade from 0.6.X->0.7.X – All the benchmarks were better – But good to kick the tires• Prototyping new Column Family – Time to write 500 million records – How many reads/second on 50GB of data
Cassandra writes fast! (duh)• Read path – Row, Key, and VFS caches – With enough data and read ops disks bottleneck• Write path – structured log writes are linear to disk-wide and fast – compaction merges sstables in background• Many threads maximizes write capability• Many threads also stops a read blocking on IO from limiting write potential
Night falls and Dr. Realtime transforms.../etc/cron.d/mr_batch_dr_realtime# turn into Mr. batch at night0 0 * * * root nodetool -h `hostname` setcompactionthroughput999#turn back into Dr. Realtime for day0 6 * * * root nodetool -h `hostname` setcompactionthroughput16Setting throughput ensures • During the day most iops are free to serve traffic • At night can rip through compactions
Mr Batch ravages data creating tombstones• If User clears cookies they vanish forever• In actuality they return as a new user• Data has very high turnover• We need to enforce retention policy on data• TTL columns do not meet our requirements :(• Cleanup daemon is a throttled range scanner• Cleanup daemon also produces histograms every cycle
Mr. Batch kills rows while you sleep
A note about different workloads• Structured log format of C* has deep implications• Many factors effect performance and disk size: • Write once data • Wide rows (many columns) • Wide rows over time (fragmented) • Application read write profile • Deletion/update percentage• LevelDB inspired compaction in 1.0 different profile then current tiered compaction
Tombstones have costs• Physically live on disk• Bloat data, index, and bloom filters• Tombstone live for a grace period and then are eligible to be removed
Caching after (major) compaction• Our case (lots of churn) major compaction shrinks data significantly• Rows fragmented over many sstables are joined• Tombstones and related data columns removed• All files should be smaller• Smaller files means better VFS caching