issue 2 hyperconverged infrastructure: the cxo view · hyperconverged infrastructure: the cxo view...
TRANSCRIPT
1
Hyperconverged Infrastructure: The CxO View Simple, cost-effective infrastructure for today’s business climate
Issue 2
In this issue
2 Welcome Fellow CxO
3 Research From Gartner: How to Determine When Hyperconverged Integrated Systems Can Replace Traditional Storage
11 911 Dispatch Center Improves Emergency Response Times with DataCore™ Hyper-converged Virtual SAN
18 About DataCore Software
2
Today’s business climate carries
a great deal of uncertainty
for companies of all sizes and
industries. Unpredictable demand
makes it difficult to focus on
long-term planning. Instead,
companies are looking for more
short-term and shifting their
investments into only their most compelling projects.
There is a strong push to simplify and reduce costs of
IT infrastructure. Server virtualization was supposed to
consolidate and simplify IT infrastructure in data centers.
But, that only “sort of happened”. Companies do have
fewer servers but they never hit the consolidation ratios
they expected. Why? In one word, performance.
Surveys show that 61% of companies have
experienced slow applications after server
virtualization with 77% pointing to I/O problems as
the culprit.
Now, with hyperconverged infrastructure, companies
have another opportunity to fulfill their vision of
consolidating and reduce the complexity of their
infrastructure. But, this will only happen if their
applications get the I/O performance they need.
DataCoreTM Hyper-converged Virtual SAN is a high-
performance, easy-to-use hyperconverged solution
that enables companies to massively consolidate
their virtualized infrastructure. Unlike other
hyperconverged vendors, DataCore is the World’s
fastest hyperconverged solution. Compared to All-
Flash Arrays (AFAs), DataCore has been proven to
have, at minimum:
Welcome Fellow CxO
■ 345% faster response time enabling faster reports,
queries and accelerating decision insights
■ 220% more IOPS enabling more powerful workloads
to be accomplished in the same timeframe
■ 230% better value on price-performance for
greater savings
What does all this performance get you? Here are
some of the benefits:
■ Faster applications
■ Consolidate your infrastructure, cutting your
CAPEX
■ Less operational costs (power, cooling and space
efficiency)
■ Less time spent managing infrastructure
This means companies can run all their applications,
even enterprise applications and databases, on the
fewest nodes of single-tier infrastructure, providing the
highest ROI by minimizing both CAPEX and OPEX.
It’s these kind of results and the advances in
performance and efficiency due to DataCore’s
revolutionary Parallel I/O technology within our
hyperconverged solution that have led to over 30,000
customer deployments globally and 96% of CxOs
surveyed stating they recommend DataCore.
Sincerely,
George Teixeira
President and CEO, Co-founder
3
Research From Gartner
How to Determine When Hyperconverged Integrated Systems Can Replace Traditional Storage
Hyperconverged integrated systems are all the IT
rage these days as vendors tout “data center in the
box” benefits. This research will help I&O leaders
distinguish the differences in HCIS and traditional
storage deployment strategies, and will provide them
with selection guidelines.
Impacts
■ Hyperconverged integrated systems’ (HCIS’)
current lack of integration with existing traditional
infrastructures causes I&O leaders to position it as
silo deployments within enterprise data centers.
■ As a result of workloads and economic analyses,
I&O leaders often deploy HCIS alongside existing
server/storage infrastructures, resulting in an
additional data center platform.
■ Most users do not go through formal HCIS
benchmarking and technical evaluation processes to
uncover differences in storage design and hardware
implementation that result in unquantified HCIS
storage efficiency, performance and ownership costs.
■ The migration of existing workloads onto HCIS is
likely to make I&O leaders to update their existing
vendor agreements, SLAs, data center design,
backup and disaster recovery strategies, staffing
and organization responsibilities.
Recommendations
■ Deploy HCIS to either consolidate all of the
midsize data center and remote office/branch
office (ROBO) workloads, or to address the specific
need for self-contained, high-impact workloads
such as VDI or virtual server infrastructure in large
enterprise data centers.
4
■ Integrate HCIS as a new platform deployed to
support well-defined, well-matched workloads and
not as a one-size-fits-all server/storage alternative.
■ Create HCIS software-defined storage (SDS)
evaluation criteria, a test plan, and an analysis
tool that assigns heavier weighting to: data
reduction ratio; performance and scalability (all
for the worst-case scenario); customer support
capabilities and the HCIS vendor’s overall
supported ecosystem.
■ Create impact analyses of switching from
traditional storage to HCIS based on vendor
proposals and bids in the areas of procurement,
facilities, networking, security, backup and disaster
recovery, and future technology deficits.
Strategic Planning Assumption
By 2019, more than 50% of the storage capacity
installed in enterprise data centers will be deployed
with SDS or HCIS architectures based on x86
commodity hardware systems, up from 10% today.
Analysis
In today’s data-driven economy, more data creation
translates immediately into increased storage
demands. In order for a business to grow rapidly,
storage needs to be able to expand in an on-demand
manner. Interest in HCIS is growing as organizations
of all sizes and market verticals seek to simplify,
speed up delivery, improve manageability and satisfy
user demand for more availability, performance and
storage capacity on tight IT budgets and with lean
resources.
Today’s Hyperconverged systems range from
reference architecture software-only products (BYOS)
to enterprise-grade hardware appliances, and
are targeted at enterprises of all sizes. By taking
advantage of the distributed scale-out nature of SDS
and elimination of single point of failure, HCIS is
designed for high availability virtualized workloads.
Vendors (see Note 1) include late-stage startups, tier-
one server and storage OEMs, and enterprise software
and hardware vendors.
When deployed correctly, for appropriate workloads
and in the right deployment model, Hyperconverged
infrastructure is a powerful architectural choice that
can transform the modern data center. This research
will explain the impact of hyperconvergence as an
alternative storage platform and how to achieve the
best possible outcomes from adopting this technology.
The first order of business is to understand how HCIS
address current pain points and deliver on simplicity,
flexibility, selectivity and economic promises.
HCIS systems have gained mind share and are being
considered as alternatives for traditional server
and storage systems in the midmarket data center,
greenfield opportunities, ROBO, data center renovation
and modernization projects for highly virtualized data
center workloads. Table 1 shows the benefits and
limitations of HCIS systems.
5
Technology IT Benefits Limitations
Traditional Servers/Storage Ability to select from broad choices
of storage and servers (selectivity)
Scale storage and compute
independently as needed (flexibility)
Integration and refresh is time and
resource consuming (economic)
Scale out is difficult (simplicity)
HCIS Seamless deployment, management
and expansion (simplicity)
Build in enterprise features such
as data reduction, backup and SSD
caching (economic)
Limited ability to independently
scale compute and storage
(flexibility)
Storage and server hardware vendor
lock-in (selectivity)Green highlighting: characteristics of strength; Red highlighting: characteristics of weakness
Source: Gartner (January 2016)
Hyperconvergence is a relative newcomer to data
center platforms modernization. Figure 1 below
shows critical differences that I&O leaders must know
before making a final decision to move away from the
traditional storage/server environment.
Impacts and Recommendations
HCIS’ current lack of integration with existing traditional infrastructures causes I&O leaders to position it as silo deployments within enterprise data centers
While Hyperconverged solutions are targeted to
flatten the IT workspace and reduce the silo effect of
different infrastructure components, the majority of
vendor implementations are not designed to integrate
with existing IT investments such a storage or server
farms, but rather to rip and replace them. That is
why HCIS is most often targeted and deployed as a
greenfield solution for a highly virtualized stack with
wide adoption in the midmarket segment, where the
integration with outside compute and storage is less
of a requirement.
HCIS data silo effect may derail deployments for
large enterprises when, instead of gaining operational
efficiency, HCIS may end up adding on another
Table 1. IT Benefits and Limitations of HCIS Versus Traditional Storage and Server Environment
platform with its own provisioning, management,
backup and DR, and capacity planning tools.
In order to avoid the data silo effect, the next
generation of HCIS will have to include some
integration capabilities with infrastructure outside of
the HCIS platform. For example, HCIS products will
have to gain the ability to ingest and control storage
on traditional storage arrays; present their own pool
of SDS for consumption by other servers in the data
center; and provision and support hybrid compute and
storage in the cloud.
Recommendations:
■ Deploy HCIS to either: consolidate all of the
midsize data center and ROBO workloads or
to address the specific need for self-contained,
high-impact workloads such as virtual desktop
infrastructure (VDI) or virtual server infrastructure
in large enterprise data centers.
■ Prioritize HCIS vendor solutions that have
integration capabilities with existing data center
investments and that will support hybrid cloud
deployments.
6
Figure 1. Critical Differences Between HCIS and the Traditional Server/Storage Approach
Source: Gartner (January 2016)
As a result of workloads and economic analyses, I&O leaders often deploy HCIS alongside existing server/storage infrastructures, resulting in an additional data center platform
As IT architects expand their design objectives
to include staff resources and ownership costs,
the appeal of integrated systems and specifically
Hyperconverged systems increases: various
integrated system implementations can include
reference architectures, integrated stack systems,
integrated infrastructure systems and HCIS. The
inherent appeal of these systems rests upon the
advantages of single vendor support, fast time
to deployment, tight HW/SW integration, ease of
provisioning and daily management, common data
services, unified life cycle management and pay as
you scale out deployment model.
HCIS has the potential to lower acquisition and
ownership costs by eliminating the expense of SAN
storage and switches, supporting data management
features such as compression and deduplication,
shrinking infrastructure delivery times, and enabling
the use of commodity servers with direct attached
disk and flash.
7
Figure 2. Impacts and Top Recommendations for Benefiting From HCIS
Source: Gartner (January 2016)
The planned service life of an HCIS system does not
have to align with server or storage system service
lives because they will often be deployed as a silo or to
support a specific project or workload.
Differences in application needs and the value maps
shown in Figure 3 indicate that cost-optimized
infrastructures will align application needs with
different technologies. Pursuing a coexistence strategy
also has the advantages of keeping competitive
pressure on traditional storage and server suppliers
to deliver aggressive pricing and effective postsales
service and support.
Deploying a HCIS solution as alternative platform
within an enterprise can enable IT to quickly satisfy
the needs of a specific business application or
workload by minimizing the testing needed to certify
its use with a variety of mission or business-critical
workloads. Examples include virtual servers, VDI
or development/testing environments. Developing
an extensible infrastructure and flexible operating
vision will help IT development by providing a viable
alternative against unwanted shadow IT. While there
are many qualitative arguments that are made in
favor of a single storage platform, the architectural
efficiencies and the benefits of maintaining a
HCIS environment might outweigh the operational
complexity and additional training they may require.
8
Figure 3. Value Map of Alternative Technologies
Source: Gartner (January 2016)
Recommendations:
■ Identify high-impact workloads that can utilize
HCIS scale-out architectures, and benefit from
HCIS low-touch deployment, ease of ongoing
management and data reduction and protection
technologies.
■ Integrate HCIS as a new platform deployed to
support well-defined, well-matched workloads and
not as a one-size-fits-all server/storage alternative.
Most users do not go through formal HCIS benchmarking and technical evaluation processes to uncover differences in storage design and hardware implementation that result in unquantified HCIS storage efficiency, performance and ownership costs
While HCIS is a relatively new deployment model,
it is expected to grow from $372 million in 2014 to
more than $5 billion by 2019 with 68% CAGR, while
remaining very fluid and fast-evolving, causing rapid
change for its product offering. All providers stress
simplicity and flexibility in various ways, but there
are subtle differences in exactly what these messages
actually translate to.
Each vendor’s HCIS implementation is likely to
exhibit unique storage efficiency, scalability and
performance profiles based on a specific workload.
HCIS decision planners need to be aware of the
wide span of HCIS offerings:
■ Hardware: Wide range of CPU, I/O optimization
hardware and SSD for caching or tiering
■ Hypervisor: Some HCIS solutions support a single
hypervisor, while others offer broader options
■ Data reduction: Some HCIS solutions offer no data
reduction, whereas others offer compression and/
or deduplication, including global deduplication
across the cluster
■ Data resiliency and efficiency: Some HCIS will
only provide data block replication, while others
can enable erasure coding, few provide the ability
to select between erasure coding and replication,
and some broader backup/disaster recovery with
application integration and file-level recovery
■ Scalability: Some HCIS clusters scale only up to eight
nodes, while others claim to scale into the hundreds
9
■ Integration: Some HCIS solutions allow integration
with existing data center infrastructure (such as
servers, storage or public cloud) while most do not
■ Data protection and availability: Some HCIS
solutions include built-in snapshots, QoS backup
and sync remote replication
There are big differences between HCIS performance, depending on hypervisor, software stack, hardware, VM density, workloads, caching and data reduction technologies.
Recommendations:
■ Include the following criteria when evaluating
HCIS: redundancy model, support and
maintenance procedures, hypervisor support, and
method of providing SDS.
■ Create HCIS software-defined storage evaluation
criteria, a test plan and an analysis tool that
assigns heavier weighting to data reduction ratio;
performance and scalability (all for the worst-case
scenario); customer support capabilities and the
HCIS vendor’s overall supported ecosystem.
■ Test HCIS solution performance under load as well
as data reduction ratios over time and at scale in
order to rightsize your cluster and finalize your
HCIS configuration.
■ Create a HCIS workload testing lab and perform
head-to-head testing by using real workloads or
storage workload generators. One example is
HCIbench, a free storage performance testing tool
for HCIS.
The migration of existing workloads onto HCIS is likely
to make I&O leaders to update their existing vendor
agreements, SLAs, data center design, backup and
disaster recovery strategies, staffing and organization
responsibilities
HCIS performance profiles and mean time between
data loss (MTBDL) will differ from existing storage/
server infrastructures. Users should identify existing
SLAs that have been made obsolete and create new
SLAs that align with HCIS capabilities. Revising
SLAs also creates an opportunity for users to cost-
optimize their operations by better aligning SLAs
with application requirements, thereby reducing the
number of situations where the infrastructure is
overdelivering against application needs. Common
measures include guaranteed I/O rates, host visible
bandwidth, response times, availability, MTBDL,
recovery point objectives (RPOs), recovery time
objectives (RTOs) and $/GB costs.
Disaster recovery schemes that rely on proprietary
HCIS-based replication technologies can only work
with other HCIS-based systems in the same family and
cannot work within existing disaster recovery schemes.
If the user has a contract with a disaster recovery
provider or colocation company, there will be contracts
to review and possible renegotiation. Possible areas
of renegotiation could include bandwidth, power and
space requirements, and the need to purchase a new
system at the disaster recovery site.
Since HCIS systems are inherently more autonomic in
their operation and require less ongoing maintenance,
their deployment could create opportunities to
revise policies and procedures that have been made
obsolete by new technologies. As a result of HCIS
10
implementation, I&O leaders will be able to reorganize
operations to improve efficiency and free budget to
reskill the organization to make it profitable rather
than a cost center.
Recommendations:
■ Build a cross-functional team that includes all
stakeholders to ensure the inclusion of current
and future storage and application requirements,
senior management support and the creation of an
effective RFP that covers the subtle consequences,
detailed in this research, of deploying an SDS
HCIS solution in the data center.
■ Engage with HCIS suppliers to profile candidate
storage workloads and create SLAs that align with
HCIS SDS capabilities and application needs.
■ Create impact analyses of switching from
traditional storage to HCIS based on vendor
proposals and bids in the areas of procurement,
facilities, networking, security, backup and disaster
recovery, and future technology deficits.
Additional research contribution and review by Arun
Chandrasekaran, Mike Cisek, Dave Russell and
George Weiss
Evidence
Evidence for this research includes more than 200
Gartner client inquiries in 2015; vendor interviews,
surveys and product demonstrations in 2014 and
2015; and customer reference surveys in 1H15.
Note 1. Sample HCIS Vendors
■ Atlantis Computing
■ Gridstor
■ Dell
■ Hitachi
■ HP
■ HTBase
■ Maxta
■ Nutanix
■ Pivot3
■ Scale Computing
■ SimpliVity
■ Springpath
■ StarWind Software
■ Stratoscale
■ VMware
Source: Source: Gartner Research, G00292287, Julia Palmer and Stanley Zaffos, 15 January 2016
11
911 Dispatch Center Improves Emergency Response Times with DataCore™ Hyper-converged Virtual SAN
Speeds up 911 Dispatch Response
Every Millisecond Counts in a 911 Call Center;
DataCore Reduces Latency Times and Makes
Critical SQL Server-based Dispatch Application Run
20X Faster
Located in Medford Oregon, Emergency
Communications of Southern Oregon (ECSO) is a
combined emergency dispatch facility and Public
Safety Answering Point (PSAP) for the 911 lines in
Jackson County Oregon. Corey Nelson, IT Manager at
ECSO, is responsible for IT at the organization. Not
only is he responsible for most of the technology
in the 911 data center, but also for almost all
Fire Department and Police Department vehicle
computers that are deployed in the field.
ECSO is a firm believer in the power of a
hyperconverged solution now that it has implemented
DataCore™ Hyper-converged Virtual SAN. Importantly, this single decision has enabled ECSO to keep using a traditional storage array by making virtual storage part of the hyperconverged infrastructure, as well as significantly increasing performance and reducing storage-related downtime.
ECSO first needed to look for a better storage
solution because its dispatch application, based on
Microsoft SQL Server, was experiencing latencies of
200 milliseconds at multiple times throughout the
day. When this application runs slow, it impacts how
fast Fire and Police can respond to an emergency. In
addition, ECSO wanted a solution to meet its key “must
haves” including better real-time mirroring, replication,
and an overall more robust storage infrastructure –
and the organization was dedicated to finding a better
alternative than its existing NetApp solution.
12
Fortunately, Nelson attended VMworld and found
DataCore. What Nelson was not thinking – even after
four intense months of looking at DataCore and
alternatives – was that a hyperconverged solution
would meet all of his tecnology and resulting business
objectives.
DataCore is deployed as hyperconverged infrastructure
using DAS or internal storage on a cluster of hosts.
DataCore Hyper-converged Virtual SAN enables users
to put the internal storage capacity of their servers
to work as a shared resource while also serving as
integrated storage architecture. Hyperconverged
systems by definition combine compute, storage,
storage networking tiers into a single unified system.
From a performance standpoint, much of the traffic
that went over the storage network could now be
eliminated and with the compute and storage co-
located faster response times were possible.
“At the time I had my first conversation with one of
DataCore’s system engineers, I was not thinking about
a hyperconverged solution,” explained Nelson. “Rather,
I was thinking about a traditional storage solution
whereby I had a separate array that handles storage
and separate hosts that would rely on that backend
storage.”
Once DataCore came onsite to ECSO and drew up
various potential solution scenarios that would meet
the organization’s infrastructure needs– focusing
specifically on a hyperconverged solution, according to
Nelson “a lightbulb went off” in his head.
“I knew then that hyperconverged was the way to
go,” emphasized Nelson. “Following that we were
able to come up with a price that suited our budget
“This product makes you think differently about storage and ultimately is the next step in virtualization. DataCore Hyper-converged Virtual SAN gives us the flexibility, reliability and performance to keep our systems running non-stop. No other products I looked at were even close to accomplishing this.”
- Corey Nelson, IT Manager, Emergency Communications of Southern Oregon
13
Customer Snapshot: Real-world Hyperconverged Scenario at ECSO
DataCore Hyper-converged Virtual SAN is perfect for environments that require high availability in a lowcost, small footprint, as well as latency-sensitive environments where the user wants to move data close to database applications, but needs to share it across a cluster of servers.
In one instance the entire ECSO building went offline because its Uninterruptible Power Supply (UPS) was being replaced. For most companies, this would mean downtime, but that is unacceptable for a 911 call center. Since Nelson had set up a back-up data center (the DR site) with DataCore, everything failed over and continued to run, despite the power outage at the primary site.
“I failed back after nine hours, and brought everything back online to the primary site,” noted Nelson. “It all worked like it was supposed to. I had zero issues from the technology side. It was great! And we stayed ‘live’ entire time. We never stopped receiving 911 calls – as that is never an option.”
– and what came next was an excellent, hassle-free
installation. I felt extremely good that DataCore Hyper-
converged Virtual SAN was the right solution for us,
which is not something I can say about the product we
had previously installed for storage management.”
NetApp was the previous storage vendor, which within
3-4 months of deployment became a huge headache
for ECSO. NetApp was paired with Dell – the server
vendor that ECSO was using prior to the NetApp
purchase. However, with a sizable investment in
NetApp, Nelson knew that he wanted to use NetApp
in some capacity. DataCore enabled him to do that to
extend the DAS capacity from each server.
The criterion that Nelson was using prior to DataCore’s
selection consisted specifically of looking for a hybrid
storage solution whereby he would incorporate some
SSD drives for performance. Nelson built out his
“selection” spreadsheet that spanned traditional
storage vendors as well as solutions that would
enable him to leverage his existing infrastructure– an
incredibly important objective since he had purchased
the NetApp technology just 18 months ago.
Performance Surges with DataCore
Prior to DataCore, performance and specifically
latency was a huge problem at ECSO – particularly
due to the NetApp array which delivered latency of
200 milliseconds on average throughout the day.
DataCore has solved the performance issues and
fixed the real-time replication issues Nelson was
previously encountering. This is because DataCore
Hyper-converged Virtual SAN speeds up response and
throughput with its innovative Parallel I/O technology
in combination with high-speed caching (using low-
latency server RAM) to keep the data close to the
applications.
The critical 911 dispatch application must interact
nearly instantly with the SQL server-based database.
Therefore during the evaluation and testing period,
understanding response and latency times were
vital criteria. To test this, Nelson ran a SQL Server
benchmark against his current environment as well as
the DataCore solution. The benchmark used a variety
of block sizes as well as a mix of random/sequential
and read/write patterns to measure the performance.
The results were, quite simply, amazing. The DataCore
Hyper-converged Virtual SAN solution was 20X faster
than his current environment, despite the fact that the
14
same nodes that generated the I/O load had to a fulfill
the requests (compared to the current environment
where separate servers generated the I/O load and all
the NetApp storage had to do was to meet the load,
which it did poorly).
“Response times are much faster. The 200 millisecond
latency has gone away now with DataCore running,”
stated Nelson. “In fact we are down to under 5
milliseconds as far as application response times at
peak load. Under normal load, the response times are
currently under one millisecond.”
Unsurpassed Storage Performance and Simplified Management using DataCore Hyper-converged Virtual SAN
Before DataCore, every storage-related task was
labor intensive at ECSO. Nelson was accessing and
reviewing documentation continuously to ensure that
any essential step concerning storage administration
was not overlooked. What became clear was that if he
went down the path of purchasing a traditional storage
SAN, it would be yet another “point” to manage.
“I wanted as few ‘panes of glass’ to manage as
possible,” commented Nelson. “Adding yet another
storage management solution to manage would just
add unnecessary complexity.”
The DataCore Hyper-converged solution was
exactly what Nelson was looking for. DataCore has
streamlined the storage management process by
automating it and enabling IT to gain visibility to
overall health and behavior of storage infrastructure
from a central console.
IT Environment At-a-Glance
■ DataCore Managed Capacity: 60 TBs
■ Are you using the auto-tiering feature? Yes
■ Number of Users: 50 internal; 250 external
■ Number of Virtual Servers and Number of Hosts: 3 hosts; 45 VMs
■ Primary Server Vendor: Dell
■ Storage Vendor(s): Dell; NetApp
■ Server Virtualization Platform: VMware ESXi 6
■ Desktop Virtualization Platform: NA
■ Hyperconverged Software: DataCore Hyper-converged Virtual SAN
DataCore Hyper-converged Virtual SAN frees Nelson
from the pain of labor-intensive storage management
and provides true hardware independence.
“DataCore has radically improved the efficiency,
performance and availability of our storage
infrastructure,” he said. “I was in the process of
purchasing new hosts, and DataCore Hyper-converged
Virtual SAN fit perfectly into the budget and plan. This
is a very unique product that can be tested in anyone’s
environment without purchasing additional hardware.”
15
As it turned out, Nelson got the “path forward’ he
wanted with DataCore Hyper-converged Virtual SAN
in that he can now rely on one pane of glass (the
DataCore management console) to manage the
storage residing on NetApp, which he just serves up
to the DataCore servers as an extension to their local
disk space.
After DataCore was implemented, NetApp was
relegated to being the low-end storage tier for use
cases such as storage archiving applications that
do not require a lot of throughput or performance.
DataCore allowed the investment in NetApp to be
protected.
The “hierarchy” of storage now at ECSO is as follows:
■ DataCore-managed flash storage comprises Tier
1 storage.
■ Tier 2 storage consists of the DataCore-managed
SAS drives.
■ Tier 3 storage is represented by the NetApp
external storage array.
With DataCore auto-tiering, all this storage is utilized
holistically to meet the performance and capacity
needs of the workloads. “Hot” data will typically
reside on tier 1, “warm” data on tier 2 and “cold”
data residing on tier 3. By automatically moving data
on a sub-LUN level basis to the tier that best matches
its performance characteristics, DataCore ensures
that each tier is used efficiently and optimally from a
performance and capacity perspective.
Delivering Real-time Data Redundancy
According to Nelson, “Now we are synchronously
mirroring to the other site. Before I may have been
doing some snapshots to the other site – but that was
timed, managed and certainly not done in realtime.
There certainly was no mirroring going on before and
latency was deplorable. Moreover, the old solution
would not allow us to failover to the backup site
without migrating the systems, therefore taking them
offline during that time. I knew that a special product
was needed to keep the systems running all of the
time. If our systems fail, it puts not only citizens but
first responders at risk.”
Two DataCore nodes reside at the primary site and
one DataCore node resides at the DR site, which is
two miles away. The DR site is connected by dark fiber
– specifically a 10-gig low-latency latency link. Both
primary site nodes mirror to the third node at the DR
site. All told, the infrastructure consists of 60 TBs of
storage including 5 TBs of SSD or flash storage.
One of Nelson’s concerns with some of his
applications was whether they could use the fiber-
linked DR site for snapshots, periodic replication or
purely synchronous mirroring.
“With DataCore, all of that works with no problem,”
said Nelson. “Protecting your data against server
outages simply by adding DataCore Hyper-
converged Virtual SAN software is easy. Within
seconds, I can migrate my two production CAD
systems over to the backup site and Dispatch is not
affected. It works great. There is zero downtime.
Nobody even knows it occurred.”
16
Nelson has brought an entire host down, while everything
was moved over to the backup host and it was “invisible”
according to him. “And then you can bring it back, which
often is the problem,” he lauded. “DataCore does it all –
from failover to failback, all seamlessly.”
Key applications are all based on SQL Server,
Exchange and Active Directory. One application ECSO
has is very unique and that is a computer-aided
dispatch application for Fire and Police. All of the data
is stored in SQL Server, but runs in a private cloud at
ECSO’s data center.
“That is really our critical application where all
information is broadcast over the network,” stated
Nelson. This gets all the Tier 1 support and it is what
everything revolves around at our site. It must always
be up-and-running.”
Better Storage Economics through Flexibility
One of the things that most appealed to Nelson about
DataCore was that if he wanted to add another server
(as has already been the case), then he could just
buy a server and turn it on – because he had already
bought enough licenses to cover a new server under
DataCore’s license terms.
“Originally I had two DataCore-powered servers
deployed and that was working just fine – and then
I added a third at a DR site just for some additional
redundancy and because I needed some more CPU
cycles,” explained Nelson. “At some point I might add
a fourth to our DR site.”
Nelson explains that adding the third node was not
particularly difficult – although he admits he is glad
a Wizard exists that he can utilize in the future when
necessary for configuring the additional nodes.
“DataCore did not even know that I added another
server because I just did it myself and turned it on,”
stated Nelson. “What is more, I did not have to buy
any specific hardware. I could have bought a bunch of
disk drives and just added those. DataCore gives me
the flexibility to build my environment how it needed to
be built.”
Summary
For ECSO, a hyperconverged solution from DataCore
accelerated their mission-critical applications while
providing huge cost-savings.
The call dispatch application, utilizing Microsoft SQL
Server, has a direct impact on the speed of Fire and
Police to respond to emergencies. With DataCore, the
tremendous performance seen during the Proofof-
Concept was matched by real-world performance in
production with peak latencies below 5 milliseconds,
whereas the application was regularly seeing latencies
of 200 milliseconds previously.
17
In addition, the organization knew that it needed new
hosts, but Nelson was prescient enough to know that
he did not want to buy new hosts without solving the
storage issue. It was during an introductory meeting
with DataCore that Nelson began to understand all of
the inherent benefits of embracing a hyperconverged
infrastructure. When the lightbulb “went off,” Nelson
realized that hyperconverged was a strategy that
could be embraced immediately by a solution readily
available from DataCore – one wherein the host and
the storage were all in one box.
“It was at that very moment that I thought – it fits our
price range and it gives us a way to use our existing
storage,” said Nelson. “It was a sheer breath of relief
once I found the solution in DataCore Hyper-converged
Virtual SAN that I had been struggling for months to
find. By implementing DataCore we would be solving
multiple issues with one purchase.”
And because of that one decision, we fixed our
storage performance issues and we upgraded our
entire infrastructure all within the budget we wanted
to spend in just a single year – rather than having to
spread out purchases to multiple years.”
About Emergency Communications of Southern Oregon
Beyond serving as a combined emergency dispatch
facility and Public Safety Answering Point (PSAP) for
the Jackson County Oregon 9-1-1 lines, ECSO is also a
regional “drop point” for emergency information that
needs to be given to Jackson and Josephine counties.
This may include severe storm warnings or notice of
a foreign enemy attack. This information is received
through the National Air Warning Alert System
(NAWAS) radio channel that covers the entire United
States. www.ecso911.com
Source: DataCore
Hyperconverged Infrastructure: The CxO View is published by DataCore. Editorial content supplied by DataCore is independent of Gartner analysis. All Gartner research is used with Gartner’s permission, and was originally published as part of Gartner’s syndicated research service available to all entitled Gartner clients. © 2016 Gartner, Inc. and/or its affiliates. All rights reserved. The use of Gartner research in this publication does not indicate Gartner’s endorsement of DataCore’s products and/or strategies. Reproduction or distribution of this publication in any form without Gartner’s prior written permission is forbidden. The information contained herein has been obtained from sources believed to be reliable. Gartner disclaims all warranties as to the accuracy, completeness or adequacy of such information. The opinions expressed herein are subject to change without notice. Although Gartner research may include a discussion of related legal issues, Gartner does not provide legal advice or services and its research should not be construed or used as such. Gartner is a public company, and its shareholders may include firms and funds that have financial interests in entities covered in Gartner research. Gartner’s Board of Directors may include senior managers of these firms or funds. Gartner research is produced independently by its research organization without input or influence from these firms, funds or their managers. For further information on the independence and integrity of Gartner research, see “Guiding Principles on Independence and Objectivity” on its website.
For additional information, please visit www.DataCore.com or email [email protected]
© 2016 DataCore Software Corporation. All Rights Reserved. DataCore, the DataCore logo and SANsymphony are trademarks or registered trademarks of DataCore Software Corporation. All other products, services and company names mentioned herein may be trademarks of their respective owners.
About DataCore SoftwareDataCore, the Data Infrastructure Software company,
is the leading provider of Software-Defined Storage
and Hyper-converged Software – harnessing today’s
powerful and cost-efficient server platforms with
Parallel I/O technology to overcome the IT industry’s
biggest problem, the I/O bottleneck, in order to
deliver unsurpassed performance, hyper-consolidation
efficiencies and cost savings. The company’s
comprehensive and flexible Software-defined Storage
and Hyper-converged Virtual SAN solutions free users
from the pain of labor-intensive storage management
and provide true independence from solutions that
cannot offer a hardware agnostic architecture.
DataCore’s storage virtualization and Parallel I/O
technology revolutionize data infrastructure and
serve as the cornerstone of the next-generation,
software-defined data center – delivering greater value,
industry-best performance, availability and simplicity.
Why DataCore and What We Do
We think differently. We innovate through software and
challenge the IT status quo.
We pioneered software-based storage virtualization.
Now, we are leading the Software-defined and Parallel
Processing revolution. Our Application-adaptive
software exploits the full potential of servers and
storage to solve data infrastructure challenges and
elevate IT to focus on the applications and services
that power their business.
DataCore parallel I/O and virtualization technologies
deliver the advantages of next generation enterprise
data centers – today – by harnessing the untapped
power of multicore servers. DataCore software
solutions revolutionize performance, cost-savings, and
productivity gains businesses can achieve from their
servers and data storage.
Contact Us
DataCore Software Corporation
Corporate Park
6300 NW 5th Way
Ft. Lauderdale, FL 33309
1 (877) 780-5111