DL Offsite

Quasar

100

This is what quasar does

extraction, normalization, chunkification, blackhole read

200

This is the normalization that transforms http://bucket.s3-aws-region.amazonaws.com to http://s3.amazonaws.com/bucket 

Quirks Normalization

300

This is the average QPS (queries per second) handled by Quasar.

50M (10-100M)

400

This is the technique used to optimize database reads in Quasar's blackhole service.

Bloom Filter

500

This SMC tier hosts Quasar's URL extraction service.

quasar_extraction

Blackhole

100

This is the concept used to define the most representative part of a URL chunk in the url detection system.

Domainish

200

This is a URL relationship used for propagation in the Blackhole system.

Redirect Chain / One Click Away

300

This team is responsible for taking care of implementing IActionDelete and the corresponding user experience.

The Enable Team

400

This is the concept used to calculate the most representative part of a url chunk

Business Type

500

These are the two Ents used to store the Blackhole data.

EntSIURLChunkStatus and EntSIURLChunkViolationType

Movie Line

100

I'm the king of the world

Titanic

200

With great power comes great responsibility.

Spider-Man (2002)

300

The cold never bothered me anyway.

Frozen

400

Hakuna Matata

The Lion King

500

Why so serious?

Dark Knight

Table

100

This table logs all IActionAbuseWriteTimeBlock

integrity_actions

200

This table contains VPV events that are sampled for integrity measurement.

integrity_central_population_vpv

300

This table logs daily blackhole mutations and blackhole mutation exceptions

si_blackhole_transition

400

This table logs all outbound clicks every day for threads

outbound_click_events_ig

500

This table contains features extracted from Integrity Crawler scrapes

ws_controller

Observation

100

This framework allows engs to create detection rules and respond to abuse in real time.

Integrity Activation Framework (IAF)

200

This system is originally designed to detect and flag potentially violating content on Meta's platforms using machine learning models.

Objectionable Content Investigation (OCIV)

300

This is the storage service for most ucc properties

tally counter

400

This is the average time that integrity crawler service takes to finish a scrape

10 minutes (5-30)

500

This is the main interface function that we use to get text from content

genFeatureMapTextArr