2017 db trends for powering real-time systems of engagement
TRANSCRIPT
1 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
2017 DB Trends for Powering Real-Time Systems of Engagement
Guest Speaker Noel Yuhanna Principal Analyst Forrester Research
Brian Bulkowski CTO and Co-Founder Aerospike, Inc.
2 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] © 2016 FORRESTER. REPRODUCTION PROHIBITED.
3 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Database Trends Powering Real-Time Systems of Engagement
Noel Yuhanna, Principal Analyst, Forrester Research
Aerospike Webinar
4 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Business growth and speed are changing the way we access data…
5 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 5 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
We Are Moving From IT to Business Technology (BT)..
Trend MIS
IT
BT
Self-service Real-time
Automation
Limited data
Digital business
Lots of data
Batch
6 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 6 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Customer Lifecycle Is More Complex Than Ever
Social media Choices
Innovation
Lower cost
Support
Appeal Blogs
Friends
Interact
7 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 7 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
TV
Direct mail
Out of home
Branch/store Call centers
Web
Social
Mobile
Customer Has Many Touch Points and Channels That Have Grown Over The Years…
8 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 8 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Systems Of Engagement Directly Touches Customers
9 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 9 © 2016 FORRESTER. REPRODUCTION PROHIBITED. 17
Age of the Customer is Driving the Need for Real-time Data Platform….
10 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 10 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Multi-dimensional View Of The Customer Drives Personalization
11 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 11 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Age of the Customer
How can you prevent members from churning?
How can you make
the customer
experience better?
How can you give the perfect customer recommendation?
12 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 12 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Customer insights
Online behavior
Social Data-as-a-service
Smart products Historical data
Scalable, low-cost, and elastic data platform
Mobile behavior and context
We Need A Data Platform To Support Enhanced Customer Insights
13 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 13 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Real-time Requirements Have Grown
› No one needs yesterday day tomorrow! › Competitive pressure – customer experience counts › Run operational reporting and analytics as then happen › New insights demand real-time data › Real-time data helps win customers › Technology is now mature to deliver real-time data platform
7
14 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 14 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Key Database trends Powering Real-time Systems of Engagement and Digital Business Transformation
Leveraging Memory and SSD NoSQL
Cloud
Database
25% adoption
20% adoption
35% adoption
15 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 15 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Mobile applications Real-time analytics
Big data analytics High-end transactional app
Packaged apps Custom apps
Scientific apps
Web apps
By 2018, 40% Enterprises Will Run Mission-critical Transactional & Operational Apps With Memory Technology
16 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 16 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Real-time Customer analytics
Systems of Engagement Business transformation
CRM Logs, Hadoop Social media Clickstream
Hybrid Memory Database
EDW Big Data DB
Analytics
Apps Apps Insights
Hybrid Memory Delivers Faster Customer Insights, and Real-time Systems of Engagement
17 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 17 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
• Prevent customers from switching .. Keep customers
• Recommendation engine • Search any data Dazzle customers
• Upsell or cross-sell new products. Sell more
• Offer the customer specific discounts. Find opportunities
..bottom line it’ll help innovate, grow your company!
How organizations are Leveraging Hybrid Memory to Support Business Transformation and SOE
18 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ] 18 © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Recommendations
• Use hybrid memory to innovate and become a disruptor, there are endless possibilities
• Intensify customer digital experiences • Insist on real-time analytics for various use
cases to gain competitive advantage
• Look closely at vendor solutions that can scale, deliver high performance are are lower cost
• Focus on tiered memory –Dram, SSD/flash • Expand more use cases to drive business growth
IT
Business
19 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
FORRESTER.COM
Thank you © 2016 FORRESTER. REPRODUCTION PROHIBITED.
Noel Yuhanna www.forrester.com Twitter: @nyuhanna
Thank you
20 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Hybrid Memory Databases for Digital Transformation
Brian Bulkowski, CTO & Co-Founder, Aerospike, Inc.
21 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Hybrid Memory Architectures enable Digital Transformation in fundamentally different ways. They deliver:
• Simplicity
• Faster Time to Market
• Business Agility
• Competitive Advantage & lower TCO
Digital Transformation – The Next Wave
22 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Traditional Architecture Has Significant Limitations
Challenges • Complex
• Maintainability
• Durability
• Consistency
• Scalability
• Cost ($)
• Data Lag
Caching Layer
Operational Database
Real-time Consumer Facing
Pricing / Inventory/Billing
Real-time Decisioning
Streaming Data
Legacy Database (Mainframe)
RDBMS Database
Transactional Systems
Enterprise Environment
Legacy RDBMS HDFS BASED
Fast speed – Consumer Scale
23 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
What’s Needed – Hybrid Memory Database
High velocity of transactions Targeted engagement requires large numbers of transactions per second
Handle huge volumes of data Must hold 100s of TBs per use case.
Non Volatile Memory (NVM) Allows you to store large volumes of data
with extremely high performance & consistency. Caching strategies routinely
fail with heavy write use cases.
Lower latency Responses to queries must consistently
be less than 5 milliseconds.
Reliability High availability – must be up 24 x 7. No single point of failure.
Easy to scale & manage Database must not require high amounts of manual intervention.
Low cost of ownership Need to reduce server “footprint” while supporting many new use cases.
Open & interoperable Need to leverage existing infrastructure & solutions.
24 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Hybrid Memory Architectures = Simplified Technology Stack
Aerospike Connectors
Legacy Database (Mainframe)
RDBMS Database
Transactional Systems
Enterprise Environment XDR
Aerospike Legacy RDBMS HDFS BASED
Powered by High Performance NoSQL
Fast speed – Consumer Scale
Hybrid Memory Database
Benefits: • Simplicity • Maintainability • Durability • Consistency • Scalability • Cost ($) • Data Lag Reduced
Real-time Consumer Facing
Pricing / Inventory/Billing
Real-time Decisioning
Streaming Data
25 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
LEGACY DATABASE (Mainframe)
XDR
App Servers
DATA WAREHOUSE/ DATA LAKE
LEGACY RDBMS HDFS BASED
BUSINESS TRANSACTIONS
Web views
( Payments/Fraud ) ( Equity Trades )
( Recommendations ) ( Real Time Billing )
High Performance Hybrid Memory DB
“REAL-TIME BIG DATA” “DECISIONING”
500 Business Trans per sec
5000 Calculations per sec
X = 2.5 M Database Transactions per sec
Hybrid Memory DB’s at Operational Scale
26 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Hybrid Memory Database - Enabling Your Digital Transformation
Powered by High Performance NoSQL
Aerospike – Hybrid Memory DB
NEXT GENERATION ARCHITECTURE• No cache required – simpler architecture!• Patented Flash Optimization – Log structured File System• Record Oriented, Schema Free NoSQL KV Store
PREDICTABLE PERFORMANCE• DRAM or Hybrid DRAM/Flash for Persistence• Stable, Low Latency and high throughput under any condition• Deployable on Bare Metal, virtualized, containerized, or Cloud
DYNAMIC CLUSTERING• Highest Uptime & Availability of any NoSQL (5 nines plus) • Automatic DB Cluster formation, healing and dynamic sharding • Cross Data Center Replication (XDR)
SMART APPS• Machine Learning• Broad language support (C/C++, Java,C#, Python, Go, Node.js, PHP)• Patented functionality, DB aware Clients• Rich API’s - Accelerated development
TCO• Optimized for Flash and DRAM• Demonstrated 10:1 price performance savings• Full Utilization ->10x reduction in servers deployed• Huge operational efficiency – “Set it and Forget it”
$
27 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Why Hybrid Memory Architecture Matters
28 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Competitive Enterprise Wins
Customer Situation Problem Aerospike Solution Customer Win
6th Largest Brokerage DB2+Gemfire cache 150 Servers growing to
1000 Single Aerospike cluster
– 12 servers - $M in TCO Savings - 99.9% of all transactions <1ms
Large Payments Firm
2 ORCL RAC clusters + Terracotta cache
System Stability & missing SLA’s
3 Aerospike Clusters – 20 Servers each
- Multi $M TCO Savings - Significant improvement in SLA’s
Fraud/Digital Identity Firm DataStax/Cassandra 168 DataStax Servers
growing to 450+ 30 Aerospike Servers –
2 clusters - $3.25M in TCO Savings over 3 years - Significant improvement in SLA’s
3rd Largest Mobile Telco
ORCL Coherence / DataStax Cassandra
Existing SOE solutions unstable & Costly
5 POC’s against Cassandra & ORCL
Coherence – all wins!
- Multi $M TCO Savings - Significant improvement in SLA’s
Enterprise Marketing
Automation Vendor
DataStax/Cassandra PostgreSQL + cache
Hundreds of cache & Cassandra Servers
Scalability challenges
Significant reduction of server footprint – global
deployment
- Multi $M TCO Savings - Significant improvement in SLA’s
30 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
AdTech – Real-Time Bidding
Challenge • Low read latency (milliseconds) • 100K to 5M operations/second • Ensure 100% uptime • Provide global data replication
Performance achieved • 1 to 6 billion cookies tracked • 5.0M auctions per second • 100ms ad rendering, 50ms real-time bidding,
1ms database access • 1.5KB median object size
Selected NoSQL • 10X fewer nodes • 10X better TCO • 20X better read latency • High throughput at low latency
Ad is Displayed
Publishers
Ad Networks & SSPs
Ad Exchanges
Demand Side Platform
Data Management Platforms
Brands Agencies Buyers
0 ms 100 ms
31 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
ATMCREDIT CARD
PROCESSING SYSTEM
FRAUD DETECTION &PROTECTION APP
ACCOUNTBEHAVIOR
ACCOUNTSTATISTICS
STATIC DATA
RULE 1 – PASSED ✔RULE 2 – PASSED ✔RULE 3 – FAILED ✗
HISTORICALDATA
RULESRULE 1RULE 2RULE 3…
Challenge ■ Overall SLA 750 ms ■ Loss of Business due to latency ■ Every Credit Card transaction requires
hundreds of DB reads/writes
Need to scale reliably ■ 10 à 100 TB ■ 10B à 100 B objects ■ 200k à I Million+ TPS
Selected NoSQL ■ Built for Flash ■ Predictable Low latency at High Throughput ■ Immediate consistency, no data loss ■ Cross data center (XDR) support ■ 20 Server Cluster ■ Dell 730xd w/ 4NVMe SSDs
Fraud Prevention
32 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Challenge ■ DB2 stores positions for 10 Million customers ■ Must update stock prices, show balances on 300 positions, process
250M transactions, 2 M updates/day ■ Running out of memory, data inconsistencies, restarts take 1 hour ■ 150 Servers -> Growing to 1000
Need to scale reliably ■ 3 à 13 TB ■ 100 à 400 Million objects ■ 200k à I Million TPS
Selected NoSQL ■ Built for Flash ■ Predictable Low latency at High Throughput ■ Immediate consistency, , no data loss ■ Cross data center (XDR) support ■ 10 Server Cluster IBM DB2
(MAINFRAME)
Read/Write
Start of Day Data Loading
End of Day Reconciliation
Query REAL-TIME DATA FEED
ACCOUNT POSITIONS
XDR
Fin Serv – Positions System of Record
33 Proprietary & Confidential | © 2016 Aerospike Inc. All rights reserved. [ ]
Challenge • Edge access to regulate traffic • Accessible using provisioning applications
(self-serve and through support personnel)
Need for Extremely High Availability, Reliably, Low latency
• > TBs of data
• 10-100M objects • 10-200K TPS
Selected NoSQL • Clustered system • Predictable low latency at high throughput
• Highly-available and reliable on failure • Cross data center (XDR) support
SOURCE DEVICE/USER DESTINATION Real-Time
Auth. QoS Billing
Request Execute Request
Real-Time Checks Config Module App
Update Device User Setting
Hot-Standby
XDR
Telco – Real-Time Billing and Charging Systems