2. We Live in an ‘Instant’ Phase of Human Evolution
Personalized & Immediate
• Our attention span is 8 sec (less than a goldfish)
• Personalized recommendations increase
conversion rates by 5.5x
• By 2020, over 50% of consumer mobile
interactions will be “hyperpersonal”
experiences
• 51% of U.S. online shoppers cite slow site
loading times as the top reason for purchase
abandonment
2
3. The ‘Instant’ Challenge
3
100 msec - The new standard for E2E application response time, under any load
50 msec - Average roundtrip internet latency
50 msec - Required round trip app response time (includes
processing & multi-DB access)
1 msec - required DB response time
DatabaseApp Servers
4. 4
The leading in-memory database platform
that supports any high performance operational,
analytics or hybrid use case
#1 Database
Used by Node.js
Developers
#1 NoSQL
on AWS
#1 Database
on Docker
Most Loved
Database
5. Vibrant Redis Community
162
CLIENTS IN 48 LANGUAGES
1000+
REDIS DOCKER LAUNCHED
PER MIN
100+
HIGHER LEVEL LIBRARIES
AND TOOLS
219
CONTRIBUTORS
45K+
STACK OVERFLOW
QUESTIONS
21.5K
REDIS GITHUB STARS
5
6. Redis Enterprise (Redise): Enterprise Grade Redis Platform
6
Database-as-a-Service
7,500+ enterprise customers
Software
200+ enterprise customers
• 6 of top Fortune 10 companies
• 3 of top 5 communications companies
Customers
• 3 of top 4 credit card issuers
• 3 of top 5 healthcare companies
7. Redis Enterprise Deployed in ALL Industries
Financial Services AdvertisingMedia
Technology Communications EducationGaming
Banks E-commerce
Business Services
Social
Travel
7
8. Redis Enterprise: Mature and Stable Platform
550K+
DATABASES RUN OVER
THREE YEARS
750
NEW DATABASES CREATED
EVERY DAY
160+
MAN-YEARS OF REDIS
ENTERPRISE TECHNOLOGY
DEVELOPMENT
1,000+
CLOUD NODE FAILURE &
OUTAGES EVENTS SURVIVED
WITHOUT DATA LOSS
13
GRANTED AND PENDING
PATENTS
100+
DEDICATED REDIS
ENGINEERS
8
9. Growth of Real-Time Data
9
“By 2025, more than a quarter of data
created in the global datasphere
will be real-time in nature” - IDC
“Between 2016 and 2019,
spending on real-time analytics
will grow 3x faster than spending on
non-real-time analytics.” - Gartner
11. Business Real-Time
“We refer to activities that occur
within 15 minutes or less of a
triggering event as being
business real time.” - Gartner
Strategic
Tactical
Operational
Automated
Operational
• Acquire a company
• Expand into a new country
• Identify target market
• Mortgage approval policy
• Hire a person
• Develop CS staff schedule
• Approve a car lease
• Adjust CS staffing dynamically
• Detect/resolve misrouted order
• Rebook an airline passenger
• Select routing for a shipment
• Pre-trade compliance verification
• Approve credit card transaction
• Algorithmic trading
Not“RealTime”“RealTime”
6 months
1 month
1 week
1 day
1 hour
15 min
1 min
1 sec
500 msec
Source: Gartner, Sept. 2016
11
16. Open Source & Enterprise Technology
Redise Node Redise Cluster
• Shared nothing cluster architecture
• Fully compatible with open source
commands & data structures
Enterprise Layer
16
Integrated Modules
17. Redis Data Structures Enable Embedded Analytics
Inline Analytics with Data Structures
Hyperlog-logs Probabilistic estimates of counts for anomaly detection
Sorted Sets Real time range analyses, top scorers, bid ranges
Sets Cardinality for fraud detection
Geospatial Indexes Location based searches
Bitmaps Real-time population counting for activity monitoring
17
18. T-digest
Rank-based statistics estimator
Redis Modules Expand Analytics Capabilities
18
Inline Analytics with Modules
Redis ML Neural Redis
Machine learning models and
model serving
Redis-cell
Rate-limiting
Rebloom
Probabilistic membership queries
Countminsketch
Approximate frequency counter
Topk
Track the top-k most frequent
elements in a stream
19. Redis Enterprise Modules Deliver Extended Capabilities
19
Operational Analytics with Modules
RediSearch
Extremely fast text-based search,
used for secondary indexing
Redis Graph
Graph query processing
Time Series
Range analyses, built-in
aggregations (min, max, sum.avg)
20. • Redis used to collect, filter, aggregate,
transform, distribute data in real-time with
sub-millisecond latency
• Redis capabilities for Fast Data Ingest with
different levels of reliability and resource
usage
- Publish/Subscribe
- Lists
- Sorted Sets as a Time Series Database
Streaming Data w/ Redis
OPTION 1: SIMPLEST
OPTION 2: RESILIENT TO CONNECTION LOSS
OPTION 3: RELIABLE, EFFICIENT
20
21. 21
The Redis Enterprise Differentiation
Integrated
ModulesAutomation
& Support
Redise Flash +
More Savings
Performance
at Scale
Built-in
Search
Active-Active
Geo
Distribution
(CRDTs) Robust
Security
HA, Durability,
DR
22. Redise : The Highest Performing Database
22
Highest Throughput at Lowest Latency
in High Volume of Writes Scenario
Least Servers Needed to
Deliver 1 Million Writes/Sec
Benchmarks performed by Avalon Consulting Group Benchmarks published in the Google blog
Serversusedtoachieve1Mwrites/sec
23. • Near-RAM performance at 80%+ lower costs
• Technology treats Flash as a RAM replacement
(or extension)
• RAM/Flash ratio can be easily configured
• Pluggable storage engine
• Available on SATA-based SSD, NVMe-based SSD,
NVDIMM like 3D XPoint/SCM on x86 and P8
platforms
Redise Flash
23
Keys & hot
values
Cold values
25. • Proven technology backed by deep
academic research
- Local latencies guaranteed with
consensus-free protocol
- Built-in conflict resolution
- Strong eventual consistency
• Multiple enhancements to make CRDTs
fully Redis compatible (CRDB)
Active - Active Geo Distribution (CRDT-Based) w/ Redis Enterprise
25
App
App
App
26. RediSearch: The World’s Fastest Search Engine
26
Better Throughput and Latency
than Elasticsearch5x
26
Multiple language support
Document and field scoring
Numeric Filtering
Stemming
Auto-suggest
Filtering by property
Ideal for (1) Secondary Index, (2) Catalog Search and (3) Text/Geo Search
27. Serving SolutionsTraining and Creation Solutions
Homegrown
Machine Learning w/ Redis
Creating a Model Serving the ModelTraining1 2 3
27
28. • Real-world challenges with Machine
Learning in Production:
– Need to serve 20k ads/sec @ 50 msec
datacenter latency
– Runs 1k campaigns -> 1k random forest
– Each forest has 15k trees with an average
depth of 7 levels
– Needs 1 trillion ops/second
• Redis-ML serves models 2000 times faster
• Cuts computing infrastructure needed by
97%
Scale Machine Learning with Redis-ML
28
2,000x faster!
msec msec
Custom App Redise
# of AWS instances 1,247xc8.xlarge 35xc8.xlarge
Reserved instances
costs
$11,448,611 $322,455
Savings N/A 97% savings
28
29. Real-time
Redise Powers the ‘Instant’ Solutions
…AND MANY MORE
29
IoT
Metering
Fraud
Mitigation Social Apps
Personal-
ization
Ecommerce
30. FUNCTIONS ESSENTIAL TO PERSONALIZATION
Personalization
DEALS FOR U
Real-Time
Analytics
Real-Time Data
Ingest
High-Speed
Transactions
Job & Queue
Management
Content
Caching
Geospatial
Data
Machine
Learning
Search
30
Dynamic
Pricing
Ads
Placement
Credit Risk
Analytics
E-Cart
Recommendations
Auto-
Discovery
Games
Experience
31. FUNCTIONS ESSENTIAL TO SOCIAL APPS
Social/Collaboration
In-Database
Analytics
Caching/User
Sessions
Fast Data
Ingest
Job & Queue
Management
Messaging/Notific
ations
JSON/Geo/Graph
31
Chat Follow Tracking Ratings Tracking Multi-player
Games
Comments
33. Metering
Rate LimitingSecurityTraffic Shaping
FUNCTIONS ESSENTIAL FOR METERING
Real-Time
Analytics
Real-Time Data
Ingest
High-Speed
Transactions
Freemium
Business Model
Tiered PricingPay As You Go
(PAYG)
Active-Active
CRDB
33
34. IoT
Smart City Transportation Industrial Agriculture Drone Management Smart Home
FUNCTIONS ESSENTIAL FOR IoT
Real-Time
Analytics
Real-Time Data
Ingest
Job & Queue
Management
Content
Caching
Geospatial
Data
Machine
Learning
SearchTime
Series
35. FUNCTIONS ESSENTIAL TO ECOMMERCE
Ecommerce
In-Database
Analytics
Caching/
User Sessions
Fast Data
Ingest
Messaging/
Notifications
JSON/Geo/Graph
35
Search Machine
Learning
High-Speed
Transactions
Inventory
Tracking
Order
Management
PromotionsPayments Loyalty ProgramAuthorization
36. Redise is Versatile
36
Covering transactional, operational and real-time analytics use cases:
Authorization
Authentication
Price Management
Advertising Bids
Messaging
Location-based Processing
User Session Management
Counting
Leaderboards
Page Ranking
Recommendation Engine
Time-series Analysis
Session Analysis
Secondary Index
Accelerated Reporting
Real-time Attribution
Search
Order History
Inventory Tracking
TRANSACTIONAL ANALYTICS OPERATIONAL
37. Home Depot
• Redis powers Home Depot’s omni-channel order
management system which is designed to process
30,000 transactions per second
• Sorted Sets and Geo data structures used for
inventory sourcing and ATP (Available to Promise)
calculation in real-time
• Hyperloglog used for counting unique customers,
sales per item, and sales per category
https://redislabs.com/webinars/implementation-patterns-
leverage-redis-turbo-charge-existing-legacy-applications/
37
38. Redfin: Redis for Real-time Analytics and More
• Redis use cases: Real-time analytics, distributed
locks for transactions, rate limiting, caching, geo
match making, search, messaging, chatting, and
more
• Real-time analytics used for:
- Error aggregation and analysis (health dashboards are
powered by Redis)
- Personalized experience to users based on location,
preferences, etc.
- Estimation of active users using hyperloglog
- Breakdown of users by city/neighborhood
38
39. Simility: Real-Time Fraud Detection
• Simility’s fraud detection platform processes billions
of transactions a day with less than 100 millisecond
latency
• Redis uses:
– As a primary datastore for Simility’s “Device Recon” module that
scores messages against fraud
– To compute unique situations using hyperloglog
– As a message broker to ingest messages from various devices
39
40. Times Internet: Redis Powers the Ad Serving Platform
• Times Internet Columbia delivers over 9 billion ad
impressions on 150+ publishers
• Redis is used for Smart Analytics – HLL, counts, real-time
performance gathering, recommendation, etc.
• 99% of ad placement requests are served in under 2
millisecond latency
40
41. Scopely: Redis for Probabilistic Analysis
• Scopely, next generation mobile entertainment
• Needs to generate on-the-fly game insights so games
can be tailored to user preferences by location,
demographics, etc
• 2.8 million events/min, 2.4 billion events/day
• Redis powers their real time system for operational
monitoring/business alerts
• Ongoing analysis of current game performance, user
engagement vs past
• Hyperloglog for estimation of different things –
examples: cheating likelihood, anomalous installs, game
play times
Analytics architecture
41
42. Imgur: Counting Billions of Image Views
• Imgur is the internet’s largest image sharing
service
• Challenge: To count and analyze over 3 billion
unique image views in real-time
• Redis replaced the previous solution based on
HBase and map-reduce
• Result:
– Faster: Processing time ~50 ms
– Cheaper: EC2 cost savings by 75%
– Simpler: No java, map-reduce, zoo keeper
42
43. Skedaddle: Pricing Model Adjustments
• Skedaddle uses Redis geospatial functionality to
generate closest pickup locations
• Redis powers pricing tool used to compare
pricing model parameters with live values
• Redis powers marketing tool to find users with
similar preferences
43
44. In the age of Personalization and Short Attention Span
Big Data means nothing
INSTANT is everything!