SlideShare a Scribd company logo
Understanding
Storage I/O Under
Load
Avi Kivity, CTO
Pavel Emelyanov, Principal Engineer
Presenter
2
Avi Kivity
● Original maintainer of Linux KVM - Kernel-based
Virtual Machine
● Co-maintainer of Seastar, ScyllaDB
● Co-founder of ScyllaDB
Presenter
3
Pavel Emelyanov
● Ex Linux kernel hacker
● Founder of the CRIU project
● Co-maintainer of Seastar & ScyllaDB
4
+ For distributed, data-intensive apps that require high
performance and low latency
+ 400+ users worldwide
+ Results
+ Comcast: Reduced P99 latencies by 95%
+ FireEye: 1500% improvement in throughput
+ Discord: Reduced C* nodes from ~140 to 6
+ iFood: 9X cost reduction vs. DynamoDB
+ Open Source, Enterprise and Cloud options
+ Fully compatible with Apache Cassandra and Amazon
DynamoDB
About ScyllaDB
1ms <1ms
10ms
1M
10M
ScyllaDB Universe of 400+ Users
400+ Companies Use ScyllaDB
Seamless experiences
across content + devices
Fast computation of flight
pricing
Corporate fleet
management
Real-time analytics
2,000,000 SKU -commerce
management
Real-time location tracking
for friends/family
Video recommendation
management
IoT for industrial
machines
Synchronize browser
properties for millions
Threat intelligence service
using JanusGraph
Real time fraud detection
across 6M
transactions/day
Uber scale, mission critical
chat & messaging app
5
Network security threat
detection
Power ~50M X1 DVRs with
billions of reqs/day
Precision healthcare via
Edison AI
Inventory hub for retail
operations
Property listings and
updates
Unified ML feature store
across the business
Cryptocurrency exchange
app
Geography-based
recommendations
Distributed storage for
distributed ledger tech
Global operations- Avon,
Body Shop + more
Predictable performance
for on sale surges
GPS-based exercise
tracking
Part I
Mixed Workloads
6
Current SSDs are Amazing
7
SSDs are Amazing
+ 6.4 GB/s read
+ 3.3 GB/s write
+ 1M read IOPS
+ 200k write IOPS
+ Often, several disks in a single server!
8
SSDs are Amazing, but not Magic
+ 6.4 GB/s read
+ OR 3.3 GB/s write
+ OR 1M read IOPS
+ OR 200k write IOPS
+ Or some kind of mix
+ But what kind of mix?!
9
+ Online transaction processing (OLTP)
+ There’s a real user at the other end
+ Maintenance workloads
+ Scaling out
+ Compaction
+ Backup
+ Analytics (OLAP)
+ Want to soak up free bandwidth, but not under a tight deadline
+ Multi-tenancy
+ Several OLTP and OLAP workloads on the same disk/data
Why mixed workloads?
10
Prevent one workload from dominating
and hurting another workload
The Challenge
11
Part II
Understanding Disks
12
Introducing Diskplorer
+ Tool to test disks at a variety of mixed workloads
+ Open source: https://github.com/scylladb/diskplorer
+ Python, fio, matplotlib
+ Fancy graphs
+ Hours of fun!
13
AWS i3en.3xlarge
14
AWS i3.3xlarge
15
AWS im4gn.4xlarge
16
GCP n2-standard-16, 8 load SSDs
17
GCP n2-standard-8, 2TB Persistent Disk
18
Clunky hard disk
19
Part III
Working with Disks
20
21
+ Equation for the “safety area” can be complex
+ Bandwidth and IOPS are not instantly measurable
Disk math troubles
Disk math (step 1)
22
23
+ Equation for the “safety area” can be complex
+ Bandwidth and IOPS are not instantly measurable
Disk math troubles
Token bucket algo (and its math)
24
Disk math (step 2)
25
Scheduler algo
26
Scheduler safety area
27
28
+ runtime – total CPU time consumed
+ virtual runtimes – total time normalized by entity “weight”
+ algorithm selects entity with smallest virtual runtime
+ long inactive entity “forgets” idle virtual runtime
CPU scheduler classes
Classes and shares
29
Scheduler real work
30
United States
2445 Faber St, Suite #200
Palo Alto, CA USA 94303
Israel
Maskit 4
Herzliya, Israel 4673304
www.scylladb.com
@scylladb
Thank You!
avi@scylladb.com
@avikivity
Ad

More Related Content

What's hot (20)

Disaggregating Ceph using NVMeoF
Disaggregating Ceph using NVMeoFDisaggregating Ceph using NVMeoF
Disaggregating Ceph using NVMeoF
ShapeBlue
 
Lec04 gpu architecture
Lec04 gpu architectureLec04 gpu architecture
Lec04 gpu architecture
Taras Zakharchenko
 
Performance Tuning EC2 Instances
Performance Tuning EC2 InstancesPerformance Tuning EC2 Instances
Performance Tuning EC2 Instances
Brendan Gregg
 
Service Function Chaining with SRv6
Service Function Chaining with SRv6Service Function Chaining with SRv6
Service Function Chaining with SRv6
Ahmed AbdelSalam
 
03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final
03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final
03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final
Gopi Krishnamurthy
 
High-Performance Networking Using eBPF, XDP, and io_uring
High-Performance Networking Using eBPF, XDP, and io_uringHigh-Performance Networking Using eBPF, XDP, and io_uring
High-Performance Networking Using eBPF, XDP, and io_uring
ScyllaDB
 
eBPF maps 101
eBPF maps 101eBPF maps 101
eBPF maps 101
SUSE Labs Taipei
 
Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling
 Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling
Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling
ScyllaDB
 
The Linux Block Layer - Built for Fast Storage
The Linux Block Layer - Built for Fast StorageThe Linux Block Layer - Built for Fast Storage
The Linux Block Layer - Built for Fast Storage
Kernel TLV
 
Performance Wins with BPF: Getting Started
Performance Wins with BPF: Getting StartedPerformance Wins with BPF: Getting Started
Performance Wins with BPF: Getting Started
Brendan Gregg
 
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack UpPushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
James Denton
 
Spark performance tuning - Maksud Ibrahimov
Spark performance tuning - Maksud IbrahimovSpark performance tuning - Maksud Ibrahimov
Spark performance tuning - Maksud Ibrahimov
Maksud Ibrahimov
 
A Peek into Google's Edge TPU
A Peek into Google's Edge TPUA Peek into Google's Edge TPU
A Peek into Google's Edge TPU
Koan-Sin Tan
 
Dpdk performance
Dpdk performanceDpdk performance
Dpdk performance
Stephen Hemminger
 
eBPF Basics
eBPF BasicseBPF Basics
eBPF Basics
Michael Kehoe
 
DPDK In Depth
DPDK In DepthDPDK In Depth
DPDK In Depth
Kernel TLV
 
Why your Spark Job is Failing
Why your Spark Job is FailingWhy your Spark Job is Failing
Why your Spark Job is Failing
DataWorks Summit
 
eBPF/XDP
eBPF/XDP eBPF/XDP
eBPF/XDP
Netronome
 
[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화
[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화
[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화
OpenStack Korea Community
 
Segment Routing Technology Deep Dive and Advanced Use Cases
Segment Routing Technology Deep Dive and Advanced Use CasesSegment Routing Technology Deep Dive and Advanced Use Cases
Segment Routing Technology Deep Dive and Advanced Use Cases
Cisco Canada
 
Disaggregating Ceph using NVMeoF
Disaggregating Ceph using NVMeoFDisaggregating Ceph using NVMeoF
Disaggregating Ceph using NVMeoF
ShapeBlue
 
Performance Tuning EC2 Instances
Performance Tuning EC2 InstancesPerformance Tuning EC2 Instances
Performance Tuning EC2 Instances
Brendan Gregg
 
Service Function Chaining with SRv6
Service Function Chaining with SRv6Service Function Chaining with SRv6
Service Function Chaining with SRv6
Ahmed AbdelSalam
 
03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final
03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final
03_03_Implementing_PCIe_ATS_in_ARM-based_SoCs_Final
Gopi Krishnamurthy
 
High-Performance Networking Using eBPF, XDP, and io_uring
High-Performance Networking Using eBPF, XDP, and io_uringHigh-Performance Networking Using eBPF, XDP, and io_uring
High-Performance Networking Using eBPF, XDP, and io_uring
ScyllaDB
 
Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling
 Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling
Scylla Summit 2022: IO Scheduling & NVMe Disk Modelling
ScyllaDB
 
The Linux Block Layer - Built for Fast Storage
The Linux Block Layer - Built for Fast StorageThe Linux Block Layer - Built for Fast Storage
The Linux Block Layer - Built for Fast Storage
Kernel TLV
 
Performance Wins with BPF: Getting Started
Performance Wins with BPF: Getting StartedPerformance Wins with BPF: Getting Started
Performance Wins with BPF: Getting Started
Brendan Gregg
 
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack UpPushing Packets - How do the ML2 Mechanism Drivers Stack Up
Pushing Packets - How do the ML2 Mechanism Drivers Stack Up
James Denton
 
Spark performance tuning - Maksud Ibrahimov
Spark performance tuning - Maksud IbrahimovSpark performance tuning - Maksud Ibrahimov
Spark performance tuning - Maksud Ibrahimov
Maksud Ibrahimov
 
A Peek into Google's Edge TPU
A Peek into Google's Edge TPUA Peek into Google's Edge TPU
A Peek into Google's Edge TPU
Koan-Sin Tan
 
Why your Spark Job is Failing
Why your Spark Job is FailingWhy your Spark Job is Failing
Why your Spark Job is Failing
DataWorks Summit
 
[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화
[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화
[OpenStack Days Korea 2016] Track1 - All flash CEPH 구성 및 최적화
OpenStack Korea Community
 
Segment Routing Technology Deep Dive and Advanced Use Cases
Segment Routing Technology Deep Dive and Advanced Use CasesSegment Routing Technology Deep Dive and Advanced Use Cases
Segment Routing Technology Deep Dive and Advanced Use Cases
Cisco Canada
 

Similar to Understanding Storage I/O Under Load (20)

Different I/O Access Methods for Linux, What We Chose for ScyllaDB, and Why
Different I/O Access Methods for Linux, What We Chose for ScyllaDB, and WhyDifferent I/O Access Methods for Linux, What We Chose for ScyllaDB, and Why
Different I/O Access Methods for Linux, What We Chose for ScyllaDB, and Why
ScyllaDB
 
Under The Hood Of A Shard-Per-Core Database Architecture
Under The Hood Of A Shard-Per-Core Database ArchitectureUnder The Hood Of A Shard-Per-Core Database Architecture
Under The Hood Of A Shard-Per-Core Database Architecture
ScyllaDB
 
Learning Rust the Hard Way for a Production Kafka + ScyllaDB Pipeline
Learning Rust the Hard Way for a Production Kafka + ScyllaDB PipelineLearning Rust the Hard Way for a Production Kafka + ScyllaDB Pipeline
Learning Rust the Hard Way for a Production Kafka + ScyllaDB Pipeline
ScyllaDB
 
Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...
Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...
Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...
DevOps.com
 
Using ScyllaDB for Extreme Scale Workloads
Using ScyllaDB for Extreme Scale WorkloadsUsing ScyllaDB for Extreme Scale Workloads
Using ScyllaDB for Extreme Scale Workloads
MarisaDelao3
 
Exploring Phantom Traffic Jams in Your Data Flows
Exploring Phantom Traffic Jams in Your Data Flows Exploring Phantom Traffic Jams in Your Data Flows
Exploring Phantom Traffic Jams in Your Data Flows
ScyllaDB
 
What’s New in ScyllaDB Open Source 5.0
What’s New in ScyllaDB Open Source 5.0What’s New in ScyllaDB Open Source 5.0
What’s New in ScyllaDB Open Source 5.0
ScyllaDB
 
Transforming the Database: Critical Innovations for Performance at Scale
Transforming the Database: Critical Innovations for Performance at ScaleTransforming the Database: Critical Innovations for Performance at Scale
Transforming the Database: Critical Innovations for Performance at Scale
ScyllaDB
 
Qnap iei partners_day_2016 1108
Qnap iei partners_day_2016 1108Qnap iei partners_day_2016 1108
Qnap iei partners_day_2016 1108
qnapivan
 
Docker Introduction
Docker IntroductionDocker Introduction
Docker Introduction
Hao Fan
 
Running a DynamoDB-compatible Database on Managed Kubernetes Services
Running a DynamoDB-compatible Database on Managed Kubernetes ServicesRunning a DynamoDB-compatible Database on Managed Kubernetes Services
Running a DynamoDB-compatible Database on Managed Kubernetes Services
ScyllaDB
 
QNAP NAS Training 2016
QNAP NAS Training 2016QNAP NAS Training 2016
QNAP NAS Training 2016
Fernando Barrientos
 
Docker intro
Docker introDocker intro
Docker intro
spiddy
 
2016-JAN-28 -- High Performance Production Databases on Ceph
2016-JAN-28 -- High Performance Production Databases on Ceph2016-JAN-28 -- High Performance Production Databases on Ceph
2016-JAN-28 -- High Performance Production Databases on Ceph
Ceph Community
 
InfiniBand for the enterprise
InfiniBand for the enterpriseInfiniBand for the enterprise
InfiniBand for the enterprise
Anas Kanzoua
 
Build Low-Latency Applications in Rust on ScyllaDB
Build Low-Latency Applications in Rust on ScyllaDBBuild Low-Latency Applications in Rust on ScyllaDB
Build Low-Latency Applications in Rust on ScyllaDB
ScyllaDB
 
PyData Boston 2013
PyData Boston 2013PyData Boston 2013
PyData Boston 2013
Travis Oliphant
 
Building a 10x More Efficient Edge Platform
Building a 10x More Efficient Edge PlatformBuilding a 10x More Efficient Edge Platform
Building a 10x More Efficient Edge Platform
ScyllaDB
 
Docker Meetup San Francisco: Radical Agility with Docker & AWS
Docker Meetup San Francisco: Radical Agility with Docker & AWSDocker Meetup San Francisco: Radical Agility with Docker & AWS
Docker Meetup San Francisco: Radical Agility with Docker & AWS
Volker Pilz
 
Build DynamoDB-Compatible Apps with Python
Build DynamoDB-Compatible Apps with PythonBuild DynamoDB-Compatible Apps with Python
Build DynamoDB-Compatible Apps with Python
ScyllaDB
 
Different I/O Access Methods for Linux, What We Chose for ScyllaDB, and Why
Different I/O Access Methods for Linux, What We Chose for ScyllaDB, and WhyDifferent I/O Access Methods for Linux, What We Chose for ScyllaDB, and Why
Different I/O Access Methods for Linux, What We Chose for ScyllaDB, and Why
ScyllaDB
 
Under The Hood Of A Shard-Per-Core Database Architecture
Under The Hood Of A Shard-Per-Core Database ArchitectureUnder The Hood Of A Shard-Per-Core Database Architecture
Under The Hood Of A Shard-Per-Core Database Architecture
ScyllaDB
 
Learning Rust the Hard Way for a Production Kafka + ScyllaDB Pipeline
Learning Rust the Hard Way for a Production Kafka + ScyllaDB PipelineLearning Rust the Hard Way for a Production Kafka + ScyllaDB Pipeline
Learning Rust the Hard Way for a Production Kafka + ScyllaDB Pipeline
ScyllaDB
 
Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...
Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...
Running a Cost-Effective DynamoDB-Compatible Database on Managed Kubernetes S...
DevOps.com
 
Using ScyllaDB for Extreme Scale Workloads
Using ScyllaDB for Extreme Scale WorkloadsUsing ScyllaDB for Extreme Scale Workloads
Using ScyllaDB for Extreme Scale Workloads
MarisaDelao3
 
Exploring Phantom Traffic Jams in Your Data Flows
Exploring Phantom Traffic Jams in Your Data Flows Exploring Phantom Traffic Jams in Your Data Flows
Exploring Phantom Traffic Jams in Your Data Flows
ScyllaDB
 
What’s New in ScyllaDB Open Source 5.0
What’s New in ScyllaDB Open Source 5.0What’s New in ScyllaDB Open Source 5.0
What’s New in ScyllaDB Open Source 5.0
ScyllaDB
 
Transforming the Database: Critical Innovations for Performance at Scale
Transforming the Database: Critical Innovations for Performance at ScaleTransforming the Database: Critical Innovations for Performance at Scale
Transforming the Database: Critical Innovations for Performance at Scale
ScyllaDB
 
Qnap iei partners_day_2016 1108
Qnap iei partners_day_2016 1108Qnap iei partners_day_2016 1108
Qnap iei partners_day_2016 1108
qnapivan
 
Docker Introduction
Docker IntroductionDocker Introduction
Docker Introduction
Hao Fan
 
Running a DynamoDB-compatible Database on Managed Kubernetes Services
Running a DynamoDB-compatible Database on Managed Kubernetes ServicesRunning a DynamoDB-compatible Database on Managed Kubernetes Services
Running a DynamoDB-compatible Database on Managed Kubernetes Services
ScyllaDB
 
Docker intro
Docker introDocker intro
Docker intro
spiddy
 
2016-JAN-28 -- High Performance Production Databases on Ceph
2016-JAN-28 -- High Performance Production Databases on Ceph2016-JAN-28 -- High Performance Production Databases on Ceph
2016-JAN-28 -- High Performance Production Databases on Ceph
Ceph Community
 
InfiniBand for the enterprise
InfiniBand for the enterpriseInfiniBand for the enterprise
InfiniBand for the enterprise
Anas Kanzoua
 
Build Low-Latency Applications in Rust on ScyllaDB
Build Low-Latency Applications in Rust on ScyllaDBBuild Low-Latency Applications in Rust on ScyllaDB
Build Low-Latency Applications in Rust on ScyllaDB
ScyllaDB
 
Building a 10x More Efficient Edge Platform
Building a 10x More Efficient Edge PlatformBuilding a 10x More Efficient Edge Platform
Building a 10x More Efficient Edge Platform
ScyllaDB
 
Docker Meetup San Francisco: Radical Agility with Docker & AWS
Docker Meetup San Francisco: Radical Agility with Docker & AWSDocker Meetup San Francisco: Radical Agility with Docker & AWS
Docker Meetup San Francisco: Radical Agility with Docker & AWS
Volker Pilz
 
Build DynamoDB-Compatible Apps with Python
Build DynamoDB-Compatible Apps with PythonBuild DynamoDB-Compatible Apps with Python
Build DynamoDB-Compatible Apps with Python
ScyllaDB
 
Ad

More from ScyllaDB (20)

Designing Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep Dive
Designing Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep DiveDesigning Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep Dive
Designing Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep Dive
ScyllaDB
 
Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...
Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...
Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...
ScyllaDB
 
Leading a High-Stakes Database Migration
Leading a High-Stakes Database MigrationLeading a High-Stakes Database Migration
Leading a High-Stakes Database Migration
ScyllaDB
 
Achieving Extreme Scale with ScyllaDB: Tips & Tradeoffs
Achieving Extreme Scale with ScyllaDB: Tips & TradeoffsAchieving Extreme Scale with ScyllaDB: Tips & Tradeoffs
Achieving Extreme Scale with ScyllaDB: Tips & Tradeoffs
ScyllaDB
 
Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...
Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...
Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...
ScyllaDB
 
How Agoda Scaled 50x Throughput with ScyllaDB by Worakarn Isaratham
How Agoda Scaled 50x Throughput with ScyllaDB by Worakarn IsarathamHow Agoda Scaled 50x Throughput with ScyllaDB by Worakarn Isaratham
How Agoda Scaled 50x Throughput with ScyllaDB by Worakarn Isaratham
ScyllaDB
 
How Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd Coleman
How Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd ColemanHow Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd Coleman
How Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd Coleman
ScyllaDB
 
ScyllaDB: 10 Years and Beyond by Dor Laor
ScyllaDB: 10 Years and Beyond by Dor LaorScyllaDB: 10 Years and Beyond by Dor Laor
ScyllaDB: 10 Years and Beyond by Dor Laor
ScyllaDB
 
Reduce Your Cloud Spend with ScyllaDB by Tzach Livyatan
Reduce Your Cloud Spend with ScyllaDB by Tzach LivyatanReduce Your Cloud Spend with ScyllaDB by Tzach Livyatan
Reduce Your Cloud Spend with ScyllaDB by Tzach Livyatan
ScyllaDB
 
Migrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence Liu
Migrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence LiuMigrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence Liu
Migrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence Liu
ScyllaDB
 
Vector Search with ScyllaDB by Szymon Wasik
Vector Search with ScyllaDB by Szymon WasikVector Search with ScyllaDB by Szymon Wasik
Vector Search with ScyllaDB by Szymon Wasik
ScyllaDB
 
Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...
Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...
Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...
ScyllaDB
 
Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...
Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...
Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...
ScyllaDB
 
Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...
Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...
Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...
ScyllaDB
 
Object Storage in ScyllaDB by Ran Regev, ScyllaDB
Object Storage in ScyllaDB by Ran Regev, ScyllaDBObject Storage in ScyllaDB by Ran Regev, ScyllaDB
Object Storage in ScyllaDB by Ran Regev, ScyllaDB
ScyllaDB
 
Lessons Learned from Building a Serverless Notifications System by Srushith R...
Lessons Learned from Building a Serverless Notifications System by Srushith R...Lessons Learned from Building a Serverless Notifications System by Srushith R...
Lessons Learned from Building a Serverless Notifications System by Srushith R...
ScyllaDB
 
A Dist Sys Programmer's Journey into AI by Piotr Sarna
A Dist Sys Programmer's Journey into AI by Piotr SarnaA Dist Sys Programmer's Journey into AI by Piotr Sarna
A Dist Sys Programmer's Journey into AI by Piotr Sarna
ScyllaDB
 
High Availability: Lessons Learned by Paul Preuveneers
High Availability: Lessons Learned by Paul PreuveneersHigh Availability: Lessons Learned by Paul Preuveneers
High Availability: Lessons Learned by Paul Preuveneers
ScyllaDB
 
How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...
How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...
How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...
ScyllaDB
 
Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...
Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...
Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...
ScyllaDB
 
Designing Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep Dive
Designing Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep DiveDesigning Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep Dive
Designing Low-Latency Systems with Rust and ScyllaDB: An Architectural Deep Dive
ScyllaDB
 
Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...
Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...
Powering a Billion Dreams: Scaling Meesho’s E-commerce Revolution with Scylla...
ScyllaDB
 
Leading a High-Stakes Database Migration
Leading a High-Stakes Database MigrationLeading a High-Stakes Database Migration
Leading a High-Stakes Database Migration
ScyllaDB
 
Achieving Extreme Scale with ScyllaDB: Tips & Tradeoffs
Achieving Extreme Scale with ScyllaDB: Tips & TradeoffsAchieving Extreme Scale with ScyllaDB: Tips & Tradeoffs
Achieving Extreme Scale with ScyllaDB: Tips & Tradeoffs
ScyllaDB
 
Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...
Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...
Securely Serving Millions of Boot Artifacts a Day by João Pedro Lima & Matt ...
ScyllaDB
 
How Agoda Scaled 50x Throughput with ScyllaDB by Worakarn Isaratham
How Agoda Scaled 50x Throughput with ScyllaDB by Worakarn IsarathamHow Agoda Scaled 50x Throughput with ScyllaDB by Worakarn Isaratham
How Agoda Scaled 50x Throughput with ScyllaDB by Worakarn Isaratham
ScyllaDB
 
How Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd Coleman
How Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd ColemanHow Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd Coleman
How Yieldmo Cut Database Costs and Cloud Dependencies Fast by Todd Coleman
ScyllaDB
 
ScyllaDB: 10 Years and Beyond by Dor Laor
ScyllaDB: 10 Years and Beyond by Dor LaorScyllaDB: 10 Years and Beyond by Dor Laor
ScyllaDB: 10 Years and Beyond by Dor Laor
ScyllaDB
 
Reduce Your Cloud Spend with ScyllaDB by Tzach Livyatan
Reduce Your Cloud Spend with ScyllaDB by Tzach LivyatanReduce Your Cloud Spend with ScyllaDB by Tzach Livyatan
Reduce Your Cloud Spend with ScyllaDB by Tzach Livyatan
ScyllaDB
 
Migrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence Liu
Migrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence LiuMigrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence Liu
Migrating 50TB Data From a Home-Grown Database to ScyllaDB, Fast by Terence Liu
ScyllaDB
 
Vector Search with ScyllaDB by Szymon Wasik
Vector Search with ScyllaDB by Szymon WasikVector Search with ScyllaDB by Szymon Wasik
Vector Search with ScyllaDB by Szymon Wasik
ScyllaDB
 
Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...
Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...
Workload Prioritization: How to Balance Multiple Workloads in a Cluster by Fe...
ScyllaDB
 
Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...
Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...
Two Leading Approaches to Data Virtualization, and Which Scales Better? by Da...
ScyllaDB
 
Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...
Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...
Scaling a Beast: Lessons from 400x Growth in a High-Stakes Financial System b...
ScyllaDB
 
Object Storage in ScyllaDB by Ran Regev, ScyllaDB
Object Storage in ScyllaDB by Ran Regev, ScyllaDBObject Storage in ScyllaDB by Ran Regev, ScyllaDB
Object Storage in ScyllaDB by Ran Regev, ScyllaDB
ScyllaDB
 
Lessons Learned from Building a Serverless Notifications System by Srushith R...
Lessons Learned from Building a Serverless Notifications System by Srushith R...Lessons Learned from Building a Serverless Notifications System by Srushith R...
Lessons Learned from Building a Serverless Notifications System by Srushith R...
ScyllaDB
 
A Dist Sys Programmer's Journey into AI by Piotr Sarna
A Dist Sys Programmer's Journey into AI by Piotr SarnaA Dist Sys Programmer's Journey into AI by Piotr Sarna
A Dist Sys Programmer's Journey into AI by Piotr Sarna
ScyllaDB
 
High Availability: Lessons Learned by Paul Preuveneers
High Availability: Lessons Learned by Paul PreuveneersHigh Availability: Lessons Learned by Paul Preuveneers
High Availability: Lessons Learned by Paul Preuveneers
ScyllaDB
 
How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...
How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...
How Natura Uses ScyllaDB and ScyllaDB Connector to Create a Real-time Data Pi...
ScyllaDB
 
Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...
Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...
Persistence Pipelines in a Processing Graph: Mutable Big Data at Salesforce b...
ScyllaDB
 
Ad

Recently uploaded (20)

Cybersecurity Identity and Access Solutions using Azure AD
Cybersecurity Identity and Access Solutions using Azure ADCybersecurity Identity and Access Solutions using Azure AD
Cybersecurity Identity and Access Solutions using Azure AD
VICTOR MAESTRE RAMIREZ
 
How Can I use the AI Hype in my Business Context?
How Can I use the AI Hype in my Business Context?How Can I use the AI Hype in my Business Context?
How Can I use the AI Hype in my Business Context?
Daniel Lehner
 
Heap, Types of Heap, Insertion and Deletion
Heap, Types of Heap, Insertion and DeletionHeap, Types of Heap, Insertion and Deletion
Heap, Types of Heap, Insertion and Deletion
Jaydeep Kale
 
Role of Data Annotation Services in AI-Powered Manufacturing
Role of Data Annotation Services in AI-Powered ManufacturingRole of Data Annotation Services in AI-Powered Manufacturing
Role of Data Annotation Services in AI-Powered Manufacturing
Andrew Leo
 
DevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptx
DevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptxDevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptx
DevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptx
Justin Reock
 
HCL Nomad Web – Best Practices und Verwaltung von Multiuser-Umgebungen
HCL Nomad Web – Best Practices und Verwaltung von Multiuser-UmgebungenHCL Nomad Web – Best Practices und Verwaltung von Multiuser-Umgebungen
HCL Nomad Web – Best Practices und Verwaltung von Multiuser-Umgebungen
panagenda
 
Increasing Retail Store Efficiency How can Planograms Save Time and Money.pptx
Increasing Retail Store Efficiency How can Planograms Save Time and Money.pptxIncreasing Retail Store Efficiency How can Planograms Save Time and Money.pptx
Increasing Retail Store Efficiency How can Planograms Save Time and Money.pptx
Anoop Ashok
 
Splunk Security Update | Public Sector Summit Germany 2025
Splunk Security Update | Public Sector Summit Germany 2025Splunk Security Update | Public Sector Summit Germany 2025
Splunk Security Update | Public Sector Summit Germany 2025
Splunk
 
Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...
Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...
Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...
Noah Loul
 
Quantum Computing Quick Research Guide by Arthur Morgan
Quantum Computing Quick Research Guide by Arthur MorganQuantum Computing Quick Research Guide by Arthur Morgan
Quantum Computing Quick Research Guide by Arthur Morgan
Arthur Morgan
 
Generative Artificial Intelligence (GenAI) in Business
Generative Artificial Intelligence (GenAI) in BusinessGenerative Artificial Intelligence (GenAI) in Business
Generative Artificial Intelligence (GenAI) in Business
Dr. Tathagat Varma
 
Electronic_Mail_Attacks-1-35.pdf by xploit
Electronic_Mail_Attacks-1-35.pdf by xploitElectronic_Mail_Attacks-1-35.pdf by xploit
Electronic_Mail_Attacks-1-35.pdf by xploit
niftliyevhuseyn
 
Into The Box Conference Keynote Day 1 (ITB2025)
Into The Box Conference Keynote Day 1 (ITB2025)Into The Box Conference Keynote Day 1 (ITB2025)
Into The Box Conference Keynote Day 1 (ITB2025)
Ortus Solutions, Corp
 
Mobile App Development Company in Saudi Arabia
Mobile App Development Company in Saudi ArabiaMobile App Development Company in Saudi Arabia
Mobile App Development Company in Saudi Arabia
Steve Jonas
 
Semantic Cultivators : The Critical Future Role to Enable AI
Semantic Cultivators : The Critical Future Role to Enable AISemantic Cultivators : The Critical Future Role to Enable AI
Semantic Cultivators : The Critical Future Role to Enable AI
artmondano
 
Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...
Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...
Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...
BookNet Canada
 
Linux Support for SMARC: How Toradex Empowers Embedded Developers
Linux Support for SMARC: How Toradex Empowers Embedded DevelopersLinux Support for SMARC: How Toradex Empowers Embedded Developers
Linux Support for SMARC: How Toradex Empowers Embedded Developers
Toradex
 
How analogue intelligence complements AI
How analogue intelligence complements AIHow analogue intelligence complements AI
How analogue intelligence complements AI
Paul Rowe
 
SAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdf
SAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdfSAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdf
SAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdf
Precisely
 
Procurement Insights Cost To Value Guide.pptx
Procurement Insights Cost To Value Guide.pptxProcurement Insights Cost To Value Guide.pptx
Procurement Insights Cost To Value Guide.pptx
Jon Hansen
 
Cybersecurity Identity and Access Solutions using Azure AD
Cybersecurity Identity and Access Solutions using Azure ADCybersecurity Identity and Access Solutions using Azure AD
Cybersecurity Identity and Access Solutions using Azure AD
VICTOR MAESTRE RAMIREZ
 
How Can I use the AI Hype in my Business Context?
How Can I use the AI Hype in my Business Context?How Can I use the AI Hype in my Business Context?
How Can I use the AI Hype in my Business Context?
Daniel Lehner
 
Heap, Types of Heap, Insertion and Deletion
Heap, Types of Heap, Insertion and DeletionHeap, Types of Heap, Insertion and Deletion
Heap, Types of Heap, Insertion and Deletion
Jaydeep Kale
 
Role of Data Annotation Services in AI-Powered Manufacturing
Role of Data Annotation Services in AI-Powered ManufacturingRole of Data Annotation Services in AI-Powered Manufacturing
Role of Data Annotation Services in AI-Powered Manufacturing
Andrew Leo
 
DevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptx
DevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptxDevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptx
DevOpsDays Atlanta 2025 - Building 10x Development Organizations.pptx
Justin Reock
 
HCL Nomad Web – Best Practices und Verwaltung von Multiuser-Umgebungen
HCL Nomad Web – Best Practices und Verwaltung von Multiuser-UmgebungenHCL Nomad Web – Best Practices und Verwaltung von Multiuser-Umgebungen
HCL Nomad Web – Best Practices und Verwaltung von Multiuser-Umgebungen
panagenda
 
Increasing Retail Store Efficiency How can Planograms Save Time and Money.pptx
Increasing Retail Store Efficiency How can Planograms Save Time and Money.pptxIncreasing Retail Store Efficiency How can Planograms Save Time and Money.pptx
Increasing Retail Store Efficiency How can Planograms Save Time and Money.pptx
Anoop Ashok
 
Splunk Security Update | Public Sector Summit Germany 2025
Splunk Security Update | Public Sector Summit Germany 2025Splunk Security Update | Public Sector Summit Germany 2025
Splunk Security Update | Public Sector Summit Germany 2025
Splunk
 
Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...
Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...
Noah Loul Shares 5 Steps to Implement AI Agents for Maximum Business Efficien...
Noah Loul
 
Quantum Computing Quick Research Guide by Arthur Morgan
Quantum Computing Quick Research Guide by Arthur MorganQuantum Computing Quick Research Guide by Arthur Morgan
Quantum Computing Quick Research Guide by Arthur Morgan
Arthur Morgan
 
Generative Artificial Intelligence (GenAI) in Business
Generative Artificial Intelligence (GenAI) in BusinessGenerative Artificial Intelligence (GenAI) in Business
Generative Artificial Intelligence (GenAI) in Business
Dr. Tathagat Varma
 
Electronic_Mail_Attacks-1-35.pdf by xploit
Electronic_Mail_Attacks-1-35.pdf by xploitElectronic_Mail_Attacks-1-35.pdf by xploit
Electronic_Mail_Attacks-1-35.pdf by xploit
niftliyevhuseyn
 
Into The Box Conference Keynote Day 1 (ITB2025)
Into The Box Conference Keynote Day 1 (ITB2025)Into The Box Conference Keynote Day 1 (ITB2025)
Into The Box Conference Keynote Day 1 (ITB2025)
Ortus Solutions, Corp
 
Mobile App Development Company in Saudi Arabia
Mobile App Development Company in Saudi ArabiaMobile App Development Company in Saudi Arabia
Mobile App Development Company in Saudi Arabia
Steve Jonas
 
Semantic Cultivators : The Critical Future Role to Enable AI
Semantic Cultivators : The Critical Future Role to Enable AISemantic Cultivators : The Critical Future Role to Enable AI
Semantic Cultivators : The Critical Future Role to Enable AI
artmondano
 
Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...
Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...
Transcript: #StandardsGoals for 2025: Standards & certification roundup - Tec...
BookNet Canada
 
Linux Support for SMARC: How Toradex Empowers Embedded Developers
Linux Support for SMARC: How Toradex Empowers Embedded DevelopersLinux Support for SMARC: How Toradex Empowers Embedded Developers
Linux Support for SMARC: How Toradex Empowers Embedded Developers
Toradex
 
How analogue intelligence complements AI
How analogue intelligence complements AIHow analogue intelligence complements AI
How analogue intelligence complements AI
Paul Rowe
 
SAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdf
SAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdfSAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdf
SAP Modernization: Maximizing the Value of Your SAP S/4HANA Migration.pdf
Precisely
 
Procurement Insights Cost To Value Guide.pptx
Procurement Insights Cost To Value Guide.pptxProcurement Insights Cost To Value Guide.pptx
Procurement Insights Cost To Value Guide.pptx
Jon Hansen
 

Understanding Storage I/O Under Load

  • 1. Understanding Storage I/O Under Load Avi Kivity, CTO Pavel Emelyanov, Principal Engineer
  • 2. Presenter 2 Avi Kivity ● Original maintainer of Linux KVM - Kernel-based Virtual Machine ● Co-maintainer of Seastar, ScyllaDB ● Co-founder of ScyllaDB
  • 3. Presenter 3 Pavel Emelyanov ● Ex Linux kernel hacker ● Founder of the CRIU project ● Co-maintainer of Seastar & ScyllaDB
  • 4. 4 + For distributed, data-intensive apps that require high performance and low latency + 400+ users worldwide + Results + Comcast: Reduced P99 latencies by 95% + FireEye: 1500% improvement in throughput + Discord: Reduced C* nodes from ~140 to 6 + iFood: 9X cost reduction vs. DynamoDB + Open Source, Enterprise and Cloud options + Fully compatible with Apache Cassandra and Amazon DynamoDB About ScyllaDB 1ms <1ms 10ms 1M 10M ScyllaDB Universe of 400+ Users
  • 5. 400+ Companies Use ScyllaDB Seamless experiences across content + devices Fast computation of flight pricing Corporate fleet management Real-time analytics 2,000,000 SKU -commerce management Real-time location tracking for friends/family Video recommendation management IoT for industrial machines Synchronize browser properties for millions Threat intelligence service using JanusGraph Real time fraud detection across 6M transactions/day Uber scale, mission critical chat & messaging app 5 Network security threat detection Power ~50M X1 DVRs with billions of reqs/day Precision healthcare via Edison AI Inventory hub for retail operations Property listings and updates Unified ML feature store across the business Cryptocurrency exchange app Geography-based recommendations Distributed storage for distributed ledger tech Global operations- Avon, Body Shop + more Predictable performance for on sale surges GPS-based exercise tracking
  • 7. Current SSDs are Amazing 7
  • 8. SSDs are Amazing + 6.4 GB/s read + 3.3 GB/s write + 1M read IOPS + 200k write IOPS + Often, several disks in a single server! 8
  • 9. SSDs are Amazing, but not Magic + 6.4 GB/s read + OR 3.3 GB/s write + OR 1M read IOPS + OR 200k write IOPS + Or some kind of mix + But what kind of mix?! 9
  • 10. + Online transaction processing (OLTP) + There’s a real user at the other end + Maintenance workloads + Scaling out + Compaction + Backup + Analytics (OLAP) + Want to soak up free bandwidth, but not under a tight deadline + Multi-tenancy + Several OLTP and OLAP workloads on the same disk/data Why mixed workloads? 10
  • 11. Prevent one workload from dominating and hurting another workload The Challenge 11
  • 13. Introducing Diskplorer + Tool to test disks at a variety of mixed workloads + Open source: https://github.com/scylladb/diskplorer + Python, fio, matplotlib + Fancy graphs + Hours of fun! 13
  • 17. GCP n2-standard-16, 8 load SSDs 17
  • 18. GCP n2-standard-8, 2TB Persistent Disk 18
  • 21. 21 + Equation for the “safety area” can be complex + Bandwidth and IOPS are not instantly measurable Disk math troubles
  • 23. 23 + Equation for the “safety area” can be complex + Bandwidth and IOPS are not instantly measurable Disk math troubles
  • 24. Token bucket algo (and its math) 24
  • 28. 28 + runtime – total CPU time consumed + virtual runtimes – total time normalized by entity “weight” + algorithm selects entity with smallest virtual runtime + long inactive entity “forgets” idle virtual runtime CPU scheduler classes
  • 31. United States 2445 Faber St, Suite #200 Palo Alto, CA USA 94303 Israel Maskit 4 Herzliya, Israel 4673304 www.scylladb.com @scylladb Thank You! [email protected] @avikivity