SlideShare a Scribd company logo
Software Defined Scale-Out NAS
Company & Product Overview
The high availability and real-;me performance
at reduced cost.
2010 Spin-Off of Nantes University
10 people
RozoFS go to market end 2013
15 deployments, 6 POC on progress
Nantes, San Francisco
Context
Source : GARTNER
New Storage Challenges
Reduce	the	infrastructure	costs
Increase capacity
Access	to	high	performance	I/O
Usual Storage got in Pain
Add disks to dedicated Hardwares
Isolated Data Islands with BoVlenecks
Usual technologies were not designed for data deluge.
Storage becomes complex and poorly performant.
Thus Scale-out has Come
Add Commodity Hardware Nodes and Aggregate Capacity
Parallel acesses to Global Namespace
=		It	scales	and	performs!
But it has its Own Pain
Data
Copies
MoreDevices means more failures...
Data must be stored 3 times to ensure reliability.
But existing codes only work for archiving...
while customers expect its usage to all data.
Replica<on Erasure	Coding
1 PB Scale-out Storage
12	Nodes	-	3	PB	-	2	Failures	
CapEx	:	180	K
6	Nodes	-	1,5	PB	-	2	Failures	
CapEx	:	90	K
Erasure Code Halves the Cost
Rozo Systems Own a New One
And we used it to create the first erasure code based software defined
scale-out NAS for datacenter intensive I/O scenarios.
Scalable to Hundreds of Petabytes
Real-;me Performance
Strong Data Protec;on
Cost Saving
Ground Breaking
Fault Tolerant Erasure Code
1, 2 or 4 seamless failures at half the cost.
The more you add nodes the more you get capacity and performance.
Scale-Out Growth Efficiency
Features
Software defined
Storage
Ready to use
packaging
No change to
Applications
Linear Increase
in Performance
Real-Time
I/O Access Strong Data
Protection
Scalable to Hundreds
of Petabytes Réduction
du TCO
Licenses	per	Node	
No	vendor	lock-in
Standard	FS	 More	Nodes,	beTer	Performance
High	IO	throughput	&		
low	latency	on	any	workload	
Innova<ve	metadata	architecture		
designed	for	high	speed	parallel	
Fault	Tolerance	in	transparent	mode	on	servers	and	disks		
Erasure	coding	applied	to	100%	of	files.	
Data	integrity	checking	&	self	healing
By	simple	addi<on	of	servers	
No	disrup<on	of	service	and	QoS		
With	linear	increase	in	performance.
No	trade-off	on	performance	&	data	protec<on.	
Improved	reliability	without	data	replica<on		
(50%	volume	savings).	
Pay-as-you-grow	mode
Op<onal	tailored	Packages
Quotas per user
and/or group
Mul; sitesGeo-Replica;on Access Key/Valor
(inode)
Posix
Ultra fast
Indexa;on
Advanced Technology
compliant
Easy Opera;on and
Maintenance
Features
Flash	ready,	up	to	100	GigE
Standard	&	simple	procedures	
Standard	management	APIs
Consul;ng
Per node license fees and/or capacity
Pre configured packages
Integra;on
Per node per year support fees
www.rozosystems.com info@rozosystems.com github.com/rozofs/rozofs
christophe.delaguerrande@rozosystems.com	
06.60.31.34.60

More Related Content

PDF
Ceph at salesforce ceph day external presentation
PDF
Red hat, inc. open storage in the enterprise 0
PPT
HDFS Issues
PDF
Ceph Days 2014 Paul Evans Slide Deck
PDF
Panasas ® Terraspark Geosciences Customer Success Story
PDF
Seagate Implementation of Dense Storage Utilizing HDDs and SSDs
ODP
HDFS presented by VIJAY
PDF
Dave Shuttleworth - Platform performance comparisons, bare metal and cloud ho...
Ceph at salesforce ceph day external presentation
Red hat, inc. open storage in the enterprise 0
HDFS Issues
Ceph Days 2014 Paul Evans Slide Deck
Panasas ® Terraspark Geosciences Customer Success Story
Seagate Implementation of Dense Storage Utilizing HDDs and SSDs
HDFS presented by VIJAY
Dave Shuttleworth - Platform performance comparisons, bare metal and cloud ho...

What's hot (8)

PDF
The Architecture of Decoupling Compute and Storage with Alluxio
PPTX
Implementation of Dense Storage Utilizing HDDs with SSDs and PCIe Flash Acc...
PDF
Decoupling Compute and Storage for Data Workloads
PDF
PPTX
test
PPTX
Webinar: Five Reasons Modern Data Centers Need Tape
PDF
Storage for big-data by Joshua Robinson
PPTX
Webinar: Three Reasons Why NAS is No Good for AI and Machine Learning
The Architecture of Decoupling Compute and Storage with Alluxio
Implementation of Dense Storage Utilizing HDDs with SSDs and PCIe Flash Acc...
Decoupling Compute and Storage for Data Workloads
test
Webinar: Five Reasons Modern Data Centers Need Tape
Storage for big-data by Joshua Robinson
Webinar: Three Reasons Why NAS is No Good for AI and Machine Learning
Ad

Similar to Overview Rozo Systems (20)

PPTX
StorageOS - 8 core principles of cloud native storage
PPTX
Webinar: NAS vs Object - Can NAS Make a Comeback?
PPTX
Migrate Existing Applications to AWS without Re-engineering
PPTX
Make a Move to AWS Now
PPTX
Barton systems Overview
PDF
Panzura & Scality - Cloud Storage made seamless - Cloud Expo New York City 2012
PPTX
Make a Move to the Azure Cloud with SoftNAS
PPTX
Red Hat Storage Day Seattle: Why Software-Defined Storage Matters
PPTX
Consolidating File Servers into the Cloud
PDF
Persistent storage in Docker
PPTX
{code} and containers
PDF
Choose the Right Container Storage for Kubernetes
PPTX
Varrow datacenter storage today and tomorrow
PPTX
Why Software-Defined Storage Matters
PPTX
Why Software-Defined Storage Matters
PPTX
Pros and Cons of Erasure Coding & Replication vs. RAID in Next-Gen Storage
PPTX
How to Guarantee High Performance for Application Data in the Cloud
PPTX
12 Architectural Requirements for Protecting Business Data in the Cloud
PPTX
Why Software-Defined Storage Matters
PPTX
Why Software-Defined Storage Matters
StorageOS - 8 core principles of cloud native storage
Webinar: NAS vs Object - Can NAS Make a Comeback?
Migrate Existing Applications to AWS without Re-engineering
Make a Move to AWS Now
Barton systems Overview
Panzura & Scality - Cloud Storage made seamless - Cloud Expo New York City 2012
Make a Move to the Azure Cloud with SoftNAS
Red Hat Storage Day Seattle: Why Software-Defined Storage Matters
Consolidating File Servers into the Cloud
Persistent storage in Docker
{code} and containers
Choose the Right Container Storage for Kubernetes
Varrow datacenter storage today and tomorrow
Why Software-Defined Storage Matters
Why Software-Defined Storage Matters
Pros and Cons of Erasure Coding & Replication vs. RAID in Next-Gen Storage
How to Guarantee High Performance for Application Data in the Cloud
12 Architectural Requirements for Protecting Business Data in the Cloud
Why Software-Defined Storage Matters
Why Software-Defined Storage Matters
Ad

Recently uploaded (20)

PPTX
Operating system designcfffgfgggggggvggggggggg
PDF
top salesforce developer skills in 2025.pdf
PPTX
CHAPTER 12 - CYBER SECURITY AND FUTURE SKILLS (1) (1).pptx
PDF
SAP S4 Hana Brochure 3 (PTS SYSTEMS AND SOLUTIONS)
PPTX
Transform Your Business with a Software ERP System
PDF
Internet Downloader Manager (IDM) Crack 6.42 Build 42 Updates Latest 2025
PDF
PTS Company Brochure 2025 (1).pdf.......
PDF
Audit Checklist Design Aligning with ISO, IATF, and Industry Standards — Omne...
PDF
Raksha Bandhan Grocery Pricing Trends in India 2025.pdf
PPTX
Agentic AI : A Practical Guide. Undersating, Implementing and Scaling Autono...
PDF
Addressing The Cult of Project Management Tools-Why Disconnected Work is Hold...
PPTX
L1 - Introduction to python Backend.pptx
PPTX
ai tools demonstartion for schools and inter college
PDF
Understanding Forklifts - TECH EHS Solution
PDF
System and Network Administraation Chapter 3
PPTX
Online Work Permit System for Fast Permit Processing
PPTX
Lecture 3: Operating Systems Introduction to Computer Hardware Systems
PPTX
VVF-Customer-Presentation2025-Ver1.9.pptx
PPTX
Agentic AI Use Case- Contract Lifecycle Management (CLM).pptx
PPTX
CHAPTER 2 - PM Management and IT Context
Operating system designcfffgfgggggggvggggggggg
top salesforce developer skills in 2025.pdf
CHAPTER 12 - CYBER SECURITY AND FUTURE SKILLS (1) (1).pptx
SAP S4 Hana Brochure 3 (PTS SYSTEMS AND SOLUTIONS)
Transform Your Business with a Software ERP System
Internet Downloader Manager (IDM) Crack 6.42 Build 42 Updates Latest 2025
PTS Company Brochure 2025 (1).pdf.......
Audit Checklist Design Aligning with ISO, IATF, and Industry Standards — Omne...
Raksha Bandhan Grocery Pricing Trends in India 2025.pdf
Agentic AI : A Practical Guide. Undersating, Implementing and Scaling Autono...
Addressing The Cult of Project Management Tools-Why Disconnected Work is Hold...
L1 - Introduction to python Backend.pptx
ai tools demonstartion for schools and inter college
Understanding Forklifts - TECH EHS Solution
System and Network Administraation Chapter 3
Online Work Permit System for Fast Permit Processing
Lecture 3: Operating Systems Introduction to Computer Hardware Systems
VVF-Customer-Presentation2025-Ver1.9.pptx
Agentic AI Use Case- Contract Lifecycle Management (CLM).pptx
CHAPTER 2 - PM Management and IT Context

Overview Rozo Systems

  • 1. Software Defined Scale-Out NAS Company & Product Overview
  • 2. The high availability and real-;me performance at reduced cost. 2010 Spin-Off of Nantes University 10 people RozoFS go to market end 2013 15 deployments, 6 POC on progress Nantes, San Francisco
  • 4. New Storage Challenges Reduce the infrastructure costs Increase capacity Access to high performance I/O
  • 5. Usual Storage got in Pain Add disks to dedicated Hardwares Isolated Data Islands with BoVlenecks Usual technologies were not designed for data deluge. Storage becomes complex and poorly performant.
  • 6. Thus Scale-out has Come Add Commodity Hardware Nodes and Aggregate Capacity Parallel acesses to Global Namespace = It scales and performs!
  • 7. But it has its Own Pain Data Copies MoreDevices means more failures... Data must be stored 3 times to ensure reliability.
  • 8. But existing codes only work for archiving... while customers expect its usage to all data. Replica<on Erasure Coding 1 PB Scale-out Storage 12 Nodes - 3 PB - 2 Failures CapEx : 180 K 6 Nodes - 1,5 PB - 2 Failures CapEx : 90 K Erasure Code Halves the Cost
  • 9. Rozo Systems Own a New One And we used it to create the first erasure code based software defined scale-out NAS for datacenter intensive I/O scenarios. Scalable to Hundreds of Petabytes Real-;me Performance Strong Data Protec;on Cost Saving
  • 10. Ground Breaking Fault Tolerant Erasure Code 1, 2 or 4 seamless failures at half the cost.
  • 11. The more you add nodes the more you get capacity and performance. Scale-Out Growth Efficiency
  • 12. Features Software defined Storage Ready to use packaging No change to Applications Linear Increase in Performance Real-Time I/O Access Strong Data Protection Scalable to Hundreds of Petabytes Réduction du TCO Licenses per Node No vendor lock-in Standard FS More Nodes, beTer Performance High IO throughput & low latency on any workload Innova<ve metadata architecture designed for high speed parallel Fault Tolerance in transparent mode on servers and disks Erasure coding applied to 100% of files. Data integrity checking & self healing By simple addi<on of servers No disrup<on of service and QoS With linear increase in performance. No trade-off on performance & data protec<on. Improved reliability without data replica<on (50% volume savings). Pay-as-you-grow mode Op<onal tailored Packages
  • 13. Quotas per user and/or group Mul; sitesGeo-Replica;on Access Key/Valor (inode) Posix Ultra fast Indexa;on Advanced Technology compliant Easy Opera;on and Maintenance Features Flash ready, up to 100 GigE Standard & simple procedures Standard management APIs
  • 14. Consul;ng Per node license fees and/or capacity Pre configured packages Integra;on Per node per year support fees