SlideShare a Scribd company logo
1 of 40
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
AWS re:INVENT
Sprinklr Uses Amazon EBS to Maximize
Its NoSQL Deployment
DAT330
N o v e m b e r , 3 0 , 2 0 1 7
J a m a l M a z h a r , H e a d o f I n f r a s t r u c t u r e a n d D e v O p s , S p r i n k l r
G o p a l a K r i s h n a P a d i d a d a k a l a , P l a t f o r m A r c h i t e c t , S p r i n k l r
A n d r e y Z a y c h i k o v , S r . S o l u t i o n s A r c h i t e c t , A W S
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Agenda
• Why and how to choose NoSQL database
• MongoDB overview & various aspects of large geo-distributed cluster
design
• Sprinklr implementation & challenges, including
• Databases Sprinklr use and their scale
• Mongo DB Deployment Topologies and Scale
• Operational challenges
• Disaster Recovery Process
• Mongo / EBS Performance stats
• Cost Savings
• Security
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Intro: What is the reason for NoSQL rise?
Complexity of data
schemes
Data volume Uptime & Latency
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
NoSQL ecosystem
• Hadoop / HBase
• Cassandra
• Scylla
• Hypertable
• Accumulo
• Cloudata
• MonetDB
• HPCC
• Apache Flink
• Splice Machine
• eXtremeDB
• ConcourseDB
• Druid
• KUDU
• Elassandra
• ArangoDB
• OrientDB
• gunDB
• MongoDB
• RethinkDB
• Couchbase Server
• CouchDB
• ToroDB
• SequoiaDB
• NosDB
• RavenDB
• MarkLogic Server
• Clusterpoint Server
• JSON ODM
• NeDB
• Terrastore
• AmisaDB:
• JasDB
• RaptorDB
• djondb
• EJDB
• densodb
• SisoDB
• SDB
• NoSQL embedded db
• ThruDB
• iBoxDB
• BergDB
• ReasonDB
• BagriDB
• Riak
• Redis
• Aerospike
• LevelDB
• RocksDB
• Berkeley DB
• GenieDB
• BangDB
• Chordless
• Scalaris
• Tokyo Cabinet / Tyrant
• Scalien
• Voldemort
• Dynomite
• KAI
• MemcacheDB
• Faircom C-Tree
• LSM
• KitaroDB
• upscaledb
• STSdb
• Tarantool/Box
• Chronicle Map
• Maxtable
• quasardb
• Pincaster
• RaptorDB
• allegro-C
• nessDB
• HyperDex
• SharedHashFile
• Symas LMDB
• Sophia
• NCache
• TayzGrid
• PickleDB
• Mnesia
• LightCloud
• Hibari
• OpenLDAP
• Genomu
• BinaryRage
• Elliptics
• DBreeze
• TreodeDB
• BoltDB
• Serenety
• Cachelot
• filejson
• InfinityDB
• SCR Siemens Common
Repository
• Neo4J
• ArangoDB, OrientDB
• Infinite Graph
• Sparksee
• TITAN
• InfoGrid
• HyperGraphDB
• GraphBase
• Trinity
• AllegroGraph
• BrightstarDB
• Meronymy
• WhiteDB
• Onyx Database
• OpenLink Virtuoso
• VertexDB
• FlockDB
• BrightstarDB
• Execom IOG
• Fallen 8
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
NoSQL options on AWS
Amazon DynamoDB Amazon S3 + Amazon
Athena
Amazon EC2 +
Amazon EBS
Amazon DynamoDB is a fast and
flexible NoSQL database service for
all applications that need consistent,
single-digit millisecond latency at
any scale. It is a fully managed cloud
database and supports both
document and key-value store
models. Its flexible data model,
reliable performance, and automatic
scaling of throughput capacity,
makes it a great fit for mobile, web
and many other applications.
Amazon Athena is an
interactive query service that
makes it easy to analyze data
in Amazon S3 using standard
SQL. Athena is serverless, so
there is no infrastructure to
manage, and you pay only
for the queries that you run.
This makes it easy for anyone
with SQL skills to quickly
analyze large-scale datasets.
Amazon Elastic Compute Cloud
(Amazon EC2) is a web service that
provides secure, resizable compute
capacity in the cloud. It is designed to
make web-scale cloud computing
easier for developers.
Amazon EC2’s simple web service
interface allows you to obtain and
configure capacity with minimal
friction.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Do you really need distributed NoSQL
clusters?
• Unstable & High Load
• No crucial requirements for data consistency
• High Uptime
• Simple data scheme without a large number of
dependencies
• No need to build reports on the fly out of the
whole amount of data
• Latency is crucial and customers are widely
distributed across multiple geos
• Close to 100% data consistency is absolute must
• Stable & predicable load without significant
changes for short period of time
• Well-defined and stable data schema
• Applications & clients can tolerate higher latency
& some deviations there
• There is a need to build reports & queries on
particular data model using full functionality of
SQL
Probably YES Definitely NOT
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Choosing NoSQL technology: Overview
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Crash course: MongoDB
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Sprinklr Overview
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Sprinklr Architecture
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Sprinklr Deployment
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Sprinklr Platform—Key technologies
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Sprinklr backend architecture evolution
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Databases in Sprinklr and their scale
1320 Nodes
947 Nodes
383 Nodes70 Nodes
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Choosing the right geo for deployment
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Designing the deployment architecture
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Example: Designing the deployment
architecture
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Application interactions with DB
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Application interactions with MongoDB
When to use L4 LB
• Geo-distributed mongos processes
• HA is critical and you experience
multiple mongos failures
Issues with configuration
ReplicaSet
Common error handling issues
• Requests throttling
• Driver errors / issues
Common issues with DB requests
• Slow requests
• Schema issues
• Connectivity
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Scale of MongoDB deployment
• 900+ servers, 900 EBS volumes
• 200+ TB of Primary data
• 250 replica sets, 20 clusters, 30
Shards on couple of big clusters, 15 TB
to 30 TB of Primary Data on each big
cluster
• ~2M mongo operations/second
across mongo deployment
• Daily 270 data volume snapshots in
less than 30 min for backup purpose.
• ~30 TB of snapshots delta sync
between east region to west region for
Disaster Recovery purpose
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Planning HA & DR
HA perspective
• Failing over on DB level &
application level
• Routing & configuration for cluster
& application
• Network is prepared for event
handling
• Handling requests increase
DR perspective
• Restoring from the backup
• Preserving data in multiple
locations
• Restoring configuration & auth
data
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
HA & DR in 2 AZ region
Master elections procedure
• Default configuration
• Election Weights
HA perspective for 2 AZ failure
• 2 major scenarios
• Local read in worst case scenario
• Proper elections weights set-up
Possible options for 2 AZ
deployments
• Classic set-up with predictable
elections
• Master + Hidden
• Remote node in a replica set
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
ReplicaSet topology
Vote :1
Primary
us-east-1b
Vote : 1
Secondary
us-east-1e
Vote : 1
Arbiter
us-east-1a
Vote : 0
HS
us-east-1b
Replication
Replication
Heartbeat
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Backups
Mongo Dump
• Incremental dumps are
not supported
• 50 GB/hour
EBS snapshots
• Incremental multiple
consistence backups
• Backup Process time <
10 Min
• Lock/stop : 2+ hours to
< 1 min
• Cost
File system copy by
using rsync
• Backup Process time
2+ hours
• Lock/stop for 2+
hours
• only one backup /
backup corruption
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Data Node Recovery
Option 1: Copy data
from EBS snapshot to
instance store
Option 2: Mongo
Sync
OR
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Migration to EBS
Migration to io1 – PIOPS
• Data node recovery reduced from 4 days to 1 - 2 hours
• Volume size
• Instance type scaling
• Oplog size decreased from 40% of volume size to 10% of
volume size.
Migration to gp2
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Disaster Recovery Process
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
MongoDB startup—How to optimize?
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Startup time speedup
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Workload—Over 2 million operations / sec
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Write workload
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Write Latency
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Read Latency
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Cost Savings
• hi1.4xlarge to r3.2xlarge and now r4.2xlarge servers
• Backups cost reduced because of EBS snapshots
• Oplog disk reduced from 40% of Volume size to 10%
of Volume size
• WiredTiger Compression
• PIOPS to gp2
• Dynamic Volume size changes
• Storage cost reduced from $155k to $53k =~ 66%
savings
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Security
EBS encryption – No
noticeable
performance impact
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Magic of cross-region data transfer with
encryption
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Choosing instance and storage type
Database implementation, data schema and access patterns should always be considered.
Compute and storage types should always be adapted to particular situation and can change
during DB lifetime.
Cassandra MongoDB
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Choosing instance and storage type
• If you want to be always cost-
effective and efficient than
deployment is a journey for you
• Consider Amazon EBS as main
option for most of the workloads
• If your performance requirements
are really high and the size of the
dataset is relatively low—consider
Amazon EC2 with ephemerals,
otherwise—go for Amazon EC2 with
Amazon EBS
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Summary—How Amazon EBS helped us
Daily incremental
consistence multiple
backups for 250+
volumes, 200 TB of
data backed up in < 30
min
DR process achieved with the acceptable RPO, RTOs
• Bandwidth capacity increase by AWS team
• Increase in limits of number of parallel snapshots we can take and transfer
• Reduced cross region sync time from 30 hours to less than 8 hours
High Availability SLA
improved
• Single digit/sub ms
Read/Write latencies
66% cost savings on
Mongo storage
• start small and
dynamically increase of
volume size as needed
• use of gp2 volumes
instead of PIOPS
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Thank you!

More Related Content

What's hot

Treeveo Pitch Deck for Presentations
Treeveo Pitch Deck for PresentationsTreeveo Pitch Deck for Presentations
Treeveo Pitch Deck for PresentationsJeroen Kemperman
 
Hadoop Migration to databricks cloud project plan.pptx
Hadoop Migration to databricks cloud project plan.pptxHadoop Migration to databricks cloud project plan.pptx
Hadoop Migration to databricks cloud project plan.pptxyashodhannn
 
The Best of Microsoft Dynamics 365 Business Central
The Best of Microsoft Dynamics 365 Business Central The Best of Microsoft Dynamics 365 Business Central
The Best of Microsoft Dynamics 365 Business Central TurnkeyTec
 
How To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics Cloud
How To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics CloudHow To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics Cloud
How To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics CloudWiiisdom
 
Rethink and Realign for Digital Transformation Success
Rethink and Realign for Digital Transformation SuccessRethink and Realign for Digital Transformation Success
Rethink and Realign for Digital Transformation SuccessPerficient, Inc.
 
CARS24 - NOAH19 Berlin
CARS24 - NOAH19 BerlinCARS24 - NOAH19 Berlin
CARS24 - NOAH19 BerlinNOAH Advisors
 
Salesforce Proposal to 3M
Salesforce Proposal to 3MSalesforce Proposal to 3M
Salesforce Proposal to 3MAnyssa Volarath
 
Business analytics in banking sector
Business analytics in banking sectorBusiness analytics in banking sector
Business analytics in banking sectorVikhilSonna
 
Cognos Analytics v11: A Closer Look
Cognos Analytics v11: A Closer Look Cognos Analytics v11: A Closer Look
Cognos Analytics v11: A Closer Look Senturus
 
Choosing technologies for a big data solution in the cloud
Choosing technologies for a big data solution in the cloudChoosing technologies for a big data solution in the cloud
Choosing technologies for a big data solution in the cloudJames Serra
 
Microsoft Dynamics 365
Microsoft Dynamics 365Microsoft Dynamics 365
Microsoft Dynamics 365IOZ AG
 

What's hot (13)

Parrot H4R Stanford 2020
Parrot H4R Stanford 2020Parrot H4R Stanford 2020
Parrot H4R Stanford 2020
 
Treeveo Pitch Deck for Presentations
Treeveo Pitch Deck for PresentationsTreeveo Pitch Deck for Presentations
Treeveo Pitch Deck for Presentations
 
Hadoop Migration to databricks cloud project plan.pptx
Hadoop Migration to databricks cloud project plan.pptxHadoop Migration to databricks cloud project plan.pptx
Hadoop Migration to databricks cloud project plan.pptx
 
The Best of Microsoft Dynamics 365 Business Central
The Best of Microsoft Dynamics 365 Business Central The Best of Microsoft Dynamics 365 Business Central
The Best of Microsoft Dynamics 365 Business Central
 
Dynamics 365
Dynamics 365Dynamics 365
Dynamics 365
 
How To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics Cloud
How To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics CloudHow To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics Cloud
How To Convert Your SAP BusinessObjects Unused Licenses To SAP Analytics Cloud
 
Rethink and Realign for Digital Transformation Success
Rethink and Realign for Digital Transformation SuccessRethink and Realign for Digital Transformation Success
Rethink and Realign for Digital Transformation Success
 
CARS24 - NOAH19 Berlin
CARS24 - NOAH19 BerlinCARS24 - NOAH19 Berlin
CARS24 - NOAH19 Berlin
 
Salesforce Proposal to 3M
Salesforce Proposal to 3MSalesforce Proposal to 3M
Salesforce Proposal to 3M
 
Business analytics in banking sector
Business analytics in banking sectorBusiness analytics in banking sector
Business analytics in banking sector
 
Cognos Analytics v11: A Closer Look
Cognos Analytics v11: A Closer Look Cognos Analytics v11: A Closer Look
Cognos Analytics v11: A Closer Look
 
Choosing technologies for a big data solution in the cloud
Choosing technologies for a big data solution in the cloudChoosing technologies for a big data solution in the cloud
Choosing technologies for a big data solution in the cloud
 
Microsoft Dynamics 365
Microsoft Dynamics 365Microsoft Dynamics 365
Microsoft Dynamics 365
 

Similar to AWS re:Invent: How Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment

DAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the CloudDAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the CloudAmazon Web Services
 
Migrating your traditional Data Warehouse to a Modern Data Lake
Migrating your traditional Data Warehouse to a Modern Data LakeMigrating your traditional Data Warehouse to a Modern Data Lake
Migrating your traditional Data Warehouse to a Modern Data LakeAmazon Web Services
 
From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...
From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...
From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...Amazon Web Services
 
Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...
Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...
Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...Amazon Web Services
 
DAT316_Report from the field on Aurora PostgreSQL Performance
DAT316_Report from the field on Aurora PostgreSQL PerformanceDAT316_Report from the field on Aurora PostgreSQL Performance
DAT316_Report from the field on Aurora PostgreSQL PerformanceAmazon Web Services
 
Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017
Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017
Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017Amazon Web Services
 
AWS Database and Analytics State of the Union
AWS Database and Analytics State of the UnionAWS Database and Analytics State of the Union
AWS Database and Analytics State of the UnionAmazon Web Services
 
Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?Amazon Web Services
 
ABD312_Deep Dive Migrating Big Data Workloads to AWS
ABD312_Deep Dive Migrating Big Data Workloads to AWSABD312_Deep Dive Migrating Big Data Workloads to AWS
ABD312_Deep Dive Migrating Big Data Workloads to AWSAmazon Web Services
 
DAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise Workloads
DAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise WorkloadsDAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise Workloads
DAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise WorkloadsAmazon Web Services
 
Heterogenous Migration with DMS and SCT: Database Week San Francisco
Heterogenous Migration with DMS and SCT: Database Week San FranciscoHeterogenous Migration with DMS and SCT: Database Week San Francisco
Heterogenous Migration with DMS and SCT: Database Week San FranciscoAmazon Web Services
 
DAT320_Moving a Galaxy into Cloud
DAT320_Moving a Galaxy into CloudDAT320_Moving a Galaxy into Cloud
DAT320_Moving a Galaxy into CloudAmazon Web Services
 
Heterogenous Migration with DMS & SCT - Michael Russo
Heterogenous Migration with DMS & SCT - Michael RussoHeterogenous Migration with DMS & SCT - Michael Russo
Heterogenous Migration with DMS & SCT - Michael RussoAmazon Web Services
 
FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...
FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...
FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...Amazon Web Services
 
SRV422 Deep Dive on AWS Database Migration Service
SRV422 Deep Dive on AWS Database Migration ServiceSRV422 Deep Dive on AWS Database Migration Service
SRV422 Deep Dive on AWS Database Migration ServiceAmazon Web Services
 
Heterogenous Migration with DMS & SCT
Heterogenous Migration with DMS & SCTHeterogenous Migration with DMS & SCT
Heterogenous Migration with DMS & SCTAmazon Web Services
 
AWS Database and Analytics State of the Union
AWS Database and Analytics State of the UnionAWS Database and Analytics State of the Union
AWS Database and Analytics State of the UnionAmazon Web Services
 
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017Amazon Web Services
 
STG206_Big Data Data Lakes and Data Oceans
STG206_Big Data Data Lakes and Data OceansSTG206_Big Data Data Lakes and Data Oceans
STG206_Big Data Data Lakes and Data OceansAmazon Web Services
 
Migrating to Amazon RDS with Database Migration Service:
Migrating to Amazon RDS with Database Migration Service:Migrating to Amazon RDS with Database Migration Service:
Migrating to Amazon RDS with Database Migration Service:Amazon Web Services
 

Similar to AWS re:Invent: How Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment (20)

DAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the CloudDAT317_Migrating Databases and Data Warehouses to the Cloud
DAT317_Migrating Databases and Data Warehouses to the Cloud
 
Migrating your traditional Data Warehouse to a Modern Data Lake
Migrating your traditional Data Warehouse to a Modern Data LakeMigrating your traditional Data Warehouse to a Modern Data Lake
Migrating your traditional Data Warehouse to a Modern Data Lake
 
From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...
From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...
From Mainframe to Microservices: Vanguard’s Move to the Cloud - ENT331 - re:I...
 
Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...
Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...
Report from the Field on the PostgreSQL-compatible Edition of Amazon Aurora -...
 
DAT316_Report from the field on Aurora PostgreSQL Performance
DAT316_Report from the field on Aurora PostgreSQL PerformanceDAT316_Report from the field on Aurora PostgreSQL Performance
DAT316_Report from the field on Aurora PostgreSQL Performance
 
Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017
Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017
Airbnb Runs on Amazon Aurora - DAT331 - re:Invent 2017
 
AWS Database and Analytics State of the Union
AWS Database and Analytics State of the UnionAWS Database and Analytics State of the Union
AWS Database and Analytics State of the Union
 
Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?Amazon Relational Database Service – How is it different to what you do today ?
Amazon Relational Database Service – How is it different to what you do today ?
 
ABD312_Deep Dive Migrating Big Data Workloads to AWS
ABD312_Deep Dive Migrating Big Data Workloads to AWSABD312_Deep Dive Migrating Big Data Workloads to AWS
ABD312_Deep Dive Migrating Big Data Workloads to AWS
 
DAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise Workloads
DAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise WorkloadsDAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise Workloads
DAT332_How Verizon is Adopting Amazon Aurora PostgreSQL for Enterprise Workloads
 
Heterogenous Migration with DMS and SCT: Database Week San Francisco
Heterogenous Migration with DMS and SCT: Database Week San FranciscoHeterogenous Migration with DMS and SCT: Database Week San Francisco
Heterogenous Migration with DMS and SCT: Database Week San Francisco
 
DAT320_Moving a Galaxy into Cloud
DAT320_Moving a Galaxy into CloudDAT320_Moving a Galaxy into Cloud
DAT320_Moving a Galaxy into Cloud
 
Heterogenous Migration with DMS & SCT - Michael Russo
Heterogenous Migration with DMS & SCT - Michael RussoHeterogenous Migration with DMS & SCT - Michael Russo
Heterogenous Migration with DMS & SCT - Michael Russo
 
FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...
FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...
FINRA's Managed Data Lake: Next-Gen Analytics in the Cloud - ENT328 - re:Inve...
 
SRV422 Deep Dive on AWS Database Migration Service
SRV422 Deep Dive on AWS Database Migration ServiceSRV422 Deep Dive on AWS Database Migration Service
SRV422 Deep Dive on AWS Database Migration Service
 
Heterogenous Migration with DMS & SCT
Heterogenous Migration with DMS & SCTHeterogenous Migration with DMS & SCT
Heterogenous Migration with DMS & SCT
 
AWS Database and Analytics State of the Union
AWS Database and Analytics State of the UnionAWS Database and Analytics State of the Union
AWS Database and Analytics State of the Union
 
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
AWS Database and Analytics State of the Union - 2017 - DAT201 - re:Invent 2017
 
STG206_Big Data Data Lakes and Data Oceans
STG206_Big Data Data Lakes and Data OceansSTG206_Big Data Data Lakes and Data Oceans
STG206_Big Data Data Lakes and Data Oceans
 
Migrating to Amazon RDS with Database Migration Service:
Migrating to Amazon RDS with Database Migration Service:Migrating to Amazon RDS with Database Migration Service:
Migrating to Amazon RDS with Database Migration Service:
 

More from Amazon Web Services

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Amazon Web Services
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Amazon Web Services
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateAmazon Web Services
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSAmazon Web Services
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Amazon Web Services
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Amazon Web Services
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...Amazon Web Services
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsAmazon Web Services
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareAmazon Web Services
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSAmazon Web Services
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAmazon Web Services
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareAmazon Web Services
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWSAmazon Web Services
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckAmazon Web Services
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without serversAmazon Web Services
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...Amazon Web Services
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceAmazon Web Services
 

More from Amazon Web Services (20)

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS Fargate
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWS
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot
 
Open banking as a service
Open banking as a serviceOpen banking as a service
Open banking as a service
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
 
Computer Vision con AWS
Computer Vision con AWSComputer Vision con AWS
Computer Vision con AWS
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatare
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e web
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWS
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch Deck
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without servers
 
Fundraising Essentials
Fundraising EssentialsFundraising Essentials
Fundraising Essentials
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container Service
 

AWS re:Invent: How Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment

  • 1. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AWS re:INVENT Sprinklr Uses Amazon EBS to Maximize Its NoSQL Deployment DAT330 N o v e m b e r , 3 0 , 2 0 1 7 J a m a l M a z h a r , H e a d o f I n f r a s t r u c t u r e a n d D e v O p s , S p r i n k l r G o p a l a K r i s h n a P a d i d a d a k a l a , P l a t f o r m A r c h i t e c t , S p r i n k l r A n d r e y Z a y c h i k o v , S r . S o l u t i o n s A r c h i t e c t , A W S
  • 2. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Agenda • Why and how to choose NoSQL database • MongoDB overview & various aspects of large geo-distributed cluster design • Sprinklr implementation & challenges, including • Databases Sprinklr use and their scale • Mongo DB Deployment Topologies and Scale • Operational challenges • Disaster Recovery Process • Mongo / EBS Performance stats • Cost Savings • Security
  • 3. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Intro: What is the reason for NoSQL rise? Complexity of data schemes Data volume Uptime & Latency
  • 4. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. NoSQL ecosystem • Hadoop / HBase • Cassandra • Scylla • Hypertable • Accumulo • Cloudata • MonetDB • HPCC • Apache Flink • Splice Machine • eXtremeDB • ConcourseDB • Druid • KUDU • Elassandra • ArangoDB • OrientDB • gunDB • MongoDB • RethinkDB • Couchbase Server • CouchDB • ToroDB • SequoiaDB • NosDB • RavenDB • MarkLogic Server • Clusterpoint Server • JSON ODM • NeDB • Terrastore • AmisaDB: • JasDB • RaptorDB • djondb • EJDB • densodb • SisoDB • SDB • NoSQL embedded db • ThruDB • iBoxDB • BergDB • ReasonDB • BagriDB • Riak • Redis • Aerospike • LevelDB • RocksDB • Berkeley DB • GenieDB • BangDB • Chordless • Scalaris • Tokyo Cabinet / Tyrant • Scalien • Voldemort • Dynomite • KAI • MemcacheDB • Faircom C-Tree • LSM • KitaroDB • upscaledb • STSdb • Tarantool/Box • Chronicle Map • Maxtable • quasardb • Pincaster • RaptorDB • allegro-C • nessDB • HyperDex • SharedHashFile • Symas LMDB • Sophia • NCache • TayzGrid • PickleDB • Mnesia • LightCloud • Hibari • OpenLDAP • Genomu • BinaryRage • Elliptics • DBreeze • TreodeDB • BoltDB • Serenety • Cachelot • filejson • InfinityDB • SCR Siemens Common Repository • Neo4J • ArangoDB, OrientDB • Infinite Graph • Sparksee • TITAN • InfoGrid • HyperGraphDB • GraphBase • Trinity • AllegroGraph • BrightstarDB • Meronymy • WhiteDB • Onyx Database • OpenLink Virtuoso • VertexDB • FlockDB • BrightstarDB • Execom IOG • Fallen 8
  • 5. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. NoSQL options on AWS Amazon DynamoDB Amazon S3 + Amazon Athena Amazon EC2 + Amazon EBS Amazon DynamoDB is a fast and flexible NoSQL database service for all applications that need consistent, single-digit millisecond latency at any scale. It is a fully managed cloud database and supports both document and key-value store models. Its flexible data model, reliable performance, and automatic scaling of throughput capacity, makes it a great fit for mobile, web and many other applications. Amazon Athena is an interactive query service that makes it easy to analyze data in Amazon S3 using standard SQL. Athena is serverless, so there is no infrastructure to manage, and you pay only for the queries that you run. This makes it easy for anyone with SQL skills to quickly analyze large-scale datasets. Amazon Elastic Compute Cloud (Amazon EC2) is a web service that provides secure, resizable compute capacity in the cloud. It is designed to make web-scale cloud computing easier for developers. Amazon EC2’s simple web service interface allows you to obtain and configure capacity with minimal friction.
  • 6. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Do you really need distributed NoSQL clusters? • Unstable & High Load • No crucial requirements for data consistency • High Uptime • Simple data scheme without a large number of dependencies • No need to build reports on the fly out of the whole amount of data • Latency is crucial and customers are widely distributed across multiple geos • Close to 100% data consistency is absolute must • Stable & predicable load without significant changes for short period of time • Well-defined and stable data schema • Applications & clients can tolerate higher latency & some deviations there • There is a need to build reports & queries on particular data model using full functionality of SQL Probably YES Definitely NOT
  • 7. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Choosing NoSQL technology: Overview
  • 8. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Crash course: MongoDB
  • 9. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Sprinklr Overview
  • 10. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Sprinklr Architecture
  • 11. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Sprinklr Deployment
  • 12. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Sprinklr Platform—Key technologies
  • 13. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Sprinklr backend architecture evolution
  • 14. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Databases in Sprinklr and their scale 1320 Nodes 947 Nodes 383 Nodes70 Nodes
  • 15. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Choosing the right geo for deployment
  • 16. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Designing the deployment architecture
  • 17. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Example: Designing the deployment architecture
  • 18. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Application interactions with DB
  • 19. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Application interactions with MongoDB When to use L4 LB • Geo-distributed mongos processes • HA is critical and you experience multiple mongos failures Issues with configuration ReplicaSet Common error handling issues • Requests throttling • Driver errors / issues Common issues with DB requests • Slow requests • Schema issues • Connectivity
  • 20. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Scale of MongoDB deployment • 900+ servers, 900 EBS volumes • 200+ TB of Primary data • 250 replica sets, 20 clusters, 30 Shards on couple of big clusters, 15 TB to 30 TB of Primary Data on each big cluster • ~2M mongo operations/second across mongo deployment • Daily 270 data volume snapshots in less than 30 min for backup purpose. • ~30 TB of snapshots delta sync between east region to west region for Disaster Recovery purpose
  • 21. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Planning HA & DR HA perspective • Failing over on DB level & application level • Routing & configuration for cluster & application • Network is prepared for event handling • Handling requests increase DR perspective • Restoring from the backup • Preserving data in multiple locations • Restoring configuration & auth data
  • 22. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. HA & DR in 2 AZ region Master elections procedure • Default configuration • Election Weights HA perspective for 2 AZ failure • 2 major scenarios • Local read in worst case scenario • Proper elections weights set-up Possible options for 2 AZ deployments • Classic set-up with predictable elections • Master + Hidden • Remote node in a replica set
  • 23. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. ReplicaSet topology Vote :1 Primary us-east-1b Vote : 1 Secondary us-east-1e Vote : 1 Arbiter us-east-1a Vote : 0 HS us-east-1b Replication Replication Heartbeat
  • 24. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Backups Mongo Dump • Incremental dumps are not supported • 50 GB/hour EBS snapshots • Incremental multiple consistence backups • Backup Process time < 10 Min • Lock/stop : 2+ hours to < 1 min • Cost File system copy by using rsync • Backup Process time 2+ hours • Lock/stop for 2+ hours • only one backup / backup corruption
  • 25. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Data Node Recovery Option 1: Copy data from EBS snapshot to instance store Option 2: Mongo Sync OR
  • 26. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Migration to EBS Migration to io1 – PIOPS • Data node recovery reduced from 4 days to 1 - 2 hours • Volume size • Instance type scaling • Oplog size decreased from 40% of volume size to 10% of volume size. Migration to gp2
  • 27. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Disaster Recovery Process
  • 28. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. MongoDB startup—How to optimize?
  • 29. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Startup time speedup
  • 30. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Workload—Over 2 million operations / sec
  • 31. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Write workload
  • 32. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Write Latency
  • 33. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Read Latency
  • 34. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Cost Savings • hi1.4xlarge to r3.2xlarge and now r4.2xlarge servers • Backups cost reduced because of EBS snapshots • Oplog disk reduced from 40% of Volume size to 10% of Volume size • WiredTiger Compression • PIOPS to gp2 • Dynamic Volume size changes • Storage cost reduced from $155k to $53k =~ 66% savings
  • 35. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Security EBS encryption – No noticeable performance impact
  • 36. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Magic of cross-region data transfer with encryption
  • 37. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Choosing instance and storage type Database implementation, data schema and access patterns should always be considered. Compute and storage types should always be adapted to particular situation and can change during DB lifetime. Cassandra MongoDB
  • 38. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Choosing instance and storage type • If you want to be always cost- effective and efficient than deployment is a journey for you • Consider Amazon EBS as main option for most of the workloads • If your performance requirements are really high and the size of the dataset is relatively low—consider Amazon EC2 with ephemerals, otherwise—go for Amazon EC2 with Amazon EBS
  • 39. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Summary—How Amazon EBS helped us Daily incremental consistence multiple backups for 250+ volumes, 200 TB of data backed up in < 30 min DR process achieved with the acceptable RPO, RTOs • Bandwidth capacity increase by AWS team • Increase in limits of number of parallel snapshots we can take and transfer • Reduced cross region sync time from 30 hours to less than 8 hours High Availability SLA improved • Single digit/sub ms Read/Write latencies 66% cost savings on Mongo storage • start small and dynamically increase of volume size as needed • use of gp2 volumes instead of PIOPS
  • 40. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Thank you!