the Top SSD Companies
trust SSD market data?
Memory Defined Software
changing face of DWPD in real products
hidden segments in the enterprise
user risk reward with big memory "flash as RAM"
the importance of being
earnest about 3DXPoint and other SSD memoryfication heresies
| in-memory cache as a
cloud service - beta from GridGain|
Editor:- June 12, 2018 - GridGain Systems
the beta release and free trials of - an in-memory cache-as-a-service that allows users to rapidly deploy
a distributed in-memory cache and access it using ANSI-99 SQL, key-value or REST
APIs. The result is in-memory computing performance in the cloud, which can be
massively scaled out and can be deployed in minutes for caching applications.
why we need cloud chips
Editor:- March 10, 2018 - "In
any computer architecture, it takes a lot more energy to fetch and schedule an
instruction than it does to execute that instruction" says Rado Danilak,
founder and CEO - Tachyum
- in his new blog -
Law Is Dying - So Where Are Its Heirs? - which among other things - shows
how the transactional costs of fetching instructions and data in classical
Editor's comments:- the needs of the cloud, coupled
with growing understanding between the tradeoffs between
and energy consumption since the widespread deployment of solid state storage
have been the inspiration for rethinking all the classical elements of computer
architecture. Some of that thinking has been rooted in the memory space but
just as significant has been a rethinking of what processors should aim to do.
Tachyum announced external funding for its Cloud Chip
last month. And
as with previous disruptive technologies - part of the warm up process for the
market - is to educate more people about how things work now so they can better
appreciate what the new technologies offer.
ioFABRIC awarded patent for latency aware software
February 27, 2018 -
it has been awarded a patent for an innovation in its Vicinity policy engine for
creating and managing data volumes based on latency requirements.
The patent protects a method for maintaining fast response times by
auto-migrating data when hardware resources are added or decommissioned, when
performance degrades due to overconsumption, or application use requirements
change. Vicinity can apply this policy even when a volume is spread over
multiple nodes and storage devices
see also:- Latency
- the SSD bookmarks,
are we ready
for infinitely faster RAM?
Microsoft acquires NASA's cloud hybridisor - Avere Systems
January 3, 2018 - Microsoft
it has agreed to acquire Avere Systems.
President and CEO - Avere Systems said - "When we started Avere Systems
in 2008, our founding ideology was to use fast, flash-based storage in the most
efficient, effective manner possible in the datacenter. Along the way, our team
of file systems experts created a technology that not only optimized critical
on-premises storage resources but also enabled enterprises to move
mission-critical, high performance application workloads to the cloud."...read
more from Ron Bianchini
Editor's comments:- There was a lot of
deep thinking in Avere. I wish them luck in the reset and recompile
SSD empowered cloud
after AFAs what's
next? - cloud adapted memory
WekaIO compares cloud storage pools to IBM FlashSystem
July 12, 2017 - - a
today emerged from stealth and
details of its cloud-native scalable file system which the company says can
deliver performance comparable to
rackmount SSDs /
Editor's comments:- The notable thing for me in this announcement
was that WekaIO uses a performance benchmark compared against an
IBM FlashSystem 900 (the
decendant of the RamSan
world's fastest storage systems from
says "Utilizing only 120 cloud compute instances with locally attached
storage, WekaIO completed 1,000 simultaneous software builds compared to 240 on
IBM's high-end FlashSystem 900. The WekaIO software utilized only 5% of the AWS
compute instance resources, leaving 95% available to run customer applications."
an ambitious positioning statement and offers users a glimpse into the kind of
performance they can get by using flash assisted cloud services. Like other
modern SSD fabric software software - "WekaIO eliminates bottlenecks and
storage silos by aggregating local SSDs inside the servers into one logical
pool, which is then presented as a single namespace to the host applications."
Walmart generates 2.5PB of analyzable data every hour
April 3, 2017 - Data Café
is a private cloud which supports business decision makers in its 20,000
stores who can access over 200 streams of internal and external data, including
40 petabytes of recent transactional data, which can be modelled, manipulated
I learned the above stats in a new case study -
Data At Walmart: How 40+ Petabytes Improves Retail Decision-Making by
business author who tells us how teams from any part of the business are invited
to bring their problems to the analytics experts and then see a solution appear
before their eyes on the nerve centre's touch screen "smart boards".
if your cloud leveraged service is down - it's your fault
March 7, 2017 - "If your business leverages AWS, and you had an outage or
degraded operation during the massive AWS outage last week, you can only blame
yourself" - says , Partner at 83North in his new blog -
Among other things Erez Ofer says - "What is happening now after each cloud
outage is a lot of learning by businesses on how to create systems that don't go
SSD stories on
cloud adapted memory systems
Editor:- January 24,
2017 - Throughout the
the data storage market we've always expected the capacity of enterprise user
memory systems to be much smaller than the capacity of all the other attached
storage in the same data processing environment.
A new blog on the
无限彩票app下载 page of
adapted memory systems - asks (among other things) if this will always be
Like many of you - I've been thinking a lot about the
evolution of memory technologies and data architectures in the past year. I
wasn't sure when would be the best time to share my thoughts about this one.
But the timing seems right now. ...read the
Nimbus awarded patent for non blocking backplane technology
December 21, 2016 - Nimbus
Data Systems today
it has been granted a patent -
- for the
non-blocking data fabric architecture which is used in its
petabyte scale SSD
architectures employed by the majority of all-flash array vendors trap flash
performance behind legacy shared bus and scale-up designs," said Thomas Isakovich,
CEO and Founder. "Now patented, Nimbus Data's Parallel Memory Architecture
overcomes the limitations of generic off-the-shelf servers, capturing the full
performance potential of all-flash technology."
Elastifile gets patent for flash-aware adaptive cloud scale data
Editor:- November 3, 2016 - Elastifile today
it has been granted a US patent (No. 9,465,558) for a method of
flash-native, collaborative data storage when running on multiple interconnected
Elastifile's technology (which is integrated in software
solutions) is aimed at the hybrid cloud market.
technology enables efficient, distributed storage across full-mesh clustered
architectures in which all nodes interact with one another across multiple sites
and clouds, in complex or constantly varying network conditions, and/or at a
scale that may encompass thousands of diverse configurations.
of the greatest challenges for private and hybrid cloud data services has been
ensuring consistent performance for distributed data writing, especially due to
noisy and mixed environments," said Ezra Hoch, chief
architect at Elastifile. "Our patented approach adaptively and efficiently
manages how and where data is written, mitigating the constantly changing
conditionsat cloud scale."
Maxta offers free 24TB version of its SDS software
Editor:- September 29, 2016 - Maxta today
the general availability of a free download of its
SDS software for qualified
organizations in the U.S., Canada and select European countries.
registrants will receive a perpetual, transferable license to a fully-featured
version of the software free of charge, enabling them to configure and deploy
a three-node HCI cluster with a maximum storage capacity of 24TB.
lowering cloud wattage with low DWPD SATA SSDs
August 4, 2016 - Although it's the faster SSD products (like
PCIe SSDs and
SSDs) which capture the attention of readers - because they show what is
possible (and after a long enough interval we see pioneering enterprise speeds
becoming commonplace at lower
prices as we're now
seeing with M.2 SSDs)
nevertheless - when it comes to where most of the SSD slots are - the workhorse
of the SSD market - in arrays, webscale and cloud - is still the simple
Well, maybe not that simple - because since about
started to see subtle power optimized and mostly read oriented (low
DWPD) SATA SSD product
lines being introduced specifically for use in dense populations in the
big market for SSD
vendors and SATA SSDs
are a low risk choice for users because there are so many competing companies
and products that ensure continuous improvements in value and quality.
new addition to this crowded market is the -
XF1230 (pdf) - a 1.9TB capacity SSD which consumes less than 5W, is
rated at 0.67 DWPD -
will ship to channel partners next month.
Weka.IO has raised over $32 million for its SDS cloud cookbook
June 8, 2016 - another new name unstealthing in the software defined storage
market is , founded in 2014,
the closing of Series B funding bringing its total capital raised to over $32
Editor's comments:- In various
cofounder and CTO - - you can see how they're progressing with their big idea of
enabling clouds and enterprises to have a single software based storage
solution with good performance, efficiency and scalability.
interesting comments here about the latency impacts of garbage collection
within the D software development environment.
Zvibel says an
infrequent (few times an hour) latency of 10mS for GC can become an infinite
wait if the kernel is stressed on memory.
of DRAM latency took many years to be widely appreciated and is what
created opportunities for PCM
and flash as DRAM (Micron
and Diablo in
SCM DIMM wars)
and also big PCIe RAM fabrics likes those from
was the spark of opportunity for Weka.IO?
(from 2014) -
- includes this comment:- "IBM, Oracle,
Microsoft, SAP, Dell and the like lost their ability to shape the future of
Pure's CEO says his legacy systems competitors are 2-3 years
behind in flash/cloud centric software
Editor:- May 26, 2016 -
reported that revenue for its recent quarter was approx $140 million, up 89%
from the year ago period.
which recaps business highlights CEO - Scott Dietzen - comments
on the nature of the competition he sees from legacy storage companies.
says - "In our view, refurbished mechanical disk-era designs from the last
century cannot fulfill the needs of the modern data center: solid-state flash
memory and cloud demand a holistic rethink. Yet the majority of FlashArray's and
all of FlashBlade's competition comes from pre-cloud disk-centric retrofits..."
Avere ranked #1 in Google's cloud partner search list
March 16 , 2016 - How well does Avere Systems (and
its virtual edge filer) work as a
Apparently very well - as Avere today
it had been named "Google Cloud Platform Technology Partner of the Year"
Plexistor releases its Software Defined Memory
January 26, 2016 - Plexistor
today announced the availability of its Software Defined Memory (SDM)
architecture for both on-premise and cloud-based deployment on EC2 for AWS.
Plexistor has a
(ppt) which outlines the launch product environment and gives indicative
Cache latency is key to side-channel attack technique which can
breach cloud server security walls
Editor:- October 29, 2015 -
Cache jitter and latencies are more than simply
quality issues - they can be the root of
The juxtaposition of these concepts in colocated
cloud servers presents
risks which were
recently by researchers at .
The research team used a combination of
techniques to first create a virtual machine on the same Amazon cloud server as
a target machine (a technique known as co-location). They then used the
co-located machine to spy on the target. By observing how it accessed
information in memory, they could determine when it was retrieving its RSA key.
Then by charting the timing of the memory access they were able to deduce the
key's actual numeric sequence. ...read the summary
OCZ does that 3rd generation SSD firmware cloud thing (but
gives it a better name)
Editor:- October 16, 2015 - It's no longer
just the newcomers to the enterprise SSD market who are doing that 3rd
generation / co-operative (whatever you want to call it) SSD controller
firmware and host stack collaboration thing.
OCZ this week
announced they're doing it too.
It's available in the
cloud oriented, read mostly SSDs). And they've got a better name for it too - "Host
Managed SSD Technology".
"Our new Saber HMS SSD, together
with a software library and API, enable for the first time (in OCZ's product
line) software orchestration of internal housekeeping tasks across large pools
of SSDs, thus overcoming performance barriers that were simply not possible to
address without this technology" said Oded Ilan, GM of OCZ's R&D
Team in Israel.
"With HMS APIs, a host can coordinate garbage
collection, log dumps, and drive geometry data" (and graphics too) in
OCZ's HMS product
Lite-On says small NVMe M.2 PCIe SSDs could be a good fit for
Editor:- August 6, 2015 - Lite-On today
a new NVMe M.2 PCIe SSD for datacenter environments.
The EP2 series
delivers R/W IOPS up to 250K /25K respectively and low latencies of 35/35 (µs).
It also has power loss protection, scalability, end-to-end data protection,
low power consumption, high endurance, sustained performance, and customized
Editor's comments:- in an earlier
release (in June
2015) about supplying a related product line to an unnamed customer
described as "one of the largest cloud service providers" Jeffrey Chang,
Lite-On's Technical Product Manager said "The
M.2 is perfect for where we
believe the future of enterprise
SSD cloud storage is going."
SuperCloud rebuilds RAID 20x faster with CoreRise PCIe SSD
July 3, 2015 - CoreRise
record breaking performance results from one of its customers -
(a well known Chinese
cloud server manufacture) based on a configuration with
in a 4U server with 2x 56Gbs
Among other things SuperCloud said its lab results showed that
RAID rebuilding was 20x
faster than without the SSD - using a RAID5 configuration of 6D+1P. While RAID
throughput was 10 to 14GB/s and 1 to 1.5 million 4KB
bath tub curve is not the most useful way of thinking about
PCIe SSD failures
Editor:- June 15, 2015 - A recently published
research study -
Study of Flash Memory Failures in the Field (pdf) - which analyzed
failure rates of PCIe
SSDs used in Facebook's infrastructure over a 4 year period - yields
some very useful insights into the user experience of large populations of
- Read disturbance errors - seem to very well managed in the enterprise SSDs
The authors said they "did not observe a statistically
significant difference in the failure rate between SSDs that have read the
most amount of data versus those that have read the least amount of data."
- Higher operational temperatures mostly led to increased failure rates,
but the effect was more pronounced for SSDs which didn't use aggressive data
throttling techniques - which could prevent runaway temperatures due to
throttling back their write performance.
- More data written by the hosts to the SSDs over time - mostly resulted in
more failures - but the authors noted that in some of the platforms studied -
more data written resulted in lower failure rates.
attributed to the fact some SSD software implementations work better at
reducing write amplification when they are exposed to more workload patterns.
contains plenty of raw data and graphs which can be a valuable resource for
SSD designers and software writers to help them understand how they can tailor
their efforts towards achieving more reliable operation. ...read
the article (pdf) See also:-
- Unlike the classic bathtub curve failure model which applies to hard drives
- SSDs can be characterized as having early an warning phase - which comes
before an early failure weed out phase of the worst drives in the population
and which precedes the onset of predicted endurance based wearout.
this aspect - a small percentage of rogue SSDs account for a disproportionately
high percentage of the total data errors in the population.
Caringo gets patent for adaptive power conservation in SDS pools
May 19, 2015 -
it has obtained a US patent for adaptive power conservation in storage
clusters. The patented technology underpins its Darkive storage management
service which (since its
in 2010) actively manages the electrical power load of its server based
storage pools according to anticipated needs.
"The access patterns
and retention requirements for enterprise data have changed considerably over
the last few years to a store-everything, always accessible approach and storage
must adapt," said , Caringo VP of Marketing. "We developed Darkive to help
organizations of any size extract every bit and watt of value while keeping
their data searchable, accessible, and protected."
the big market impact
of SSD dark matter
another design win for Seagate's Nytro in China无限彩票app下载 cloud market
March 12, 2015 -
mentioned high capacity and
low cost among the reasons for selecting Seagate's
(pdf) (PCIe SSD)
as components to build the low latency SSD infrastructure of its cloud
services for the China无限彩票app下载 market - in a
comments:- who are the new cloud companies in China无限彩票app下载?
China无限彩票app下载's Cloud Innovators - a blog by Charlie Dai, Principal
the big market impact
of SSD dark matter
the Top 10 Hyperscale Sites
Editor:- December 10,
2014 - IT Brand
Pulse recently published a new list -
on its publication -
So you won't be surprised IT Brand Pulse says - "These hyperscale
sites are some of the biggest data centers around the world..." ...read
70% of raw enterprise storage capacity will be in hyperscale
datacenters by 2016
Editor:- December 9, 2014 - In its
press release today - IDC says - "By 2016,
hyperscale datacenters will house more than 50% of raw compute capacity and 70%
of raw storage capacity worldwide, becoming the primary consumers/adopters of
new compute and storage technologies."
Efficiency is important for web scale users - says Coho
October 9, 2014 -
as a file system - a web scale case study - a new blog by Andy Warfield , cofounder
and CTO - Coho Data
- made very interesting reading for me - as much for revealing the
authoritative approach taken in Andy's systematic analysis - as for the object
of his discussion (Facebook's storage architecture).
It reveals useful
insights into the architectural thinking and value judgments of Coho's
technology - and is not simply another retelling of the Facebook infrastructure
it you may get different things out of it - because it's rich in raw
enterprise ideas related to
dark matter users.
All of which makes it hard to pick out any single quote. But here are 2.
- re -
match between enterprise products and user needs
says - "In the past, enterprise hardware has had a pretty hands-off
relationship with the vendor that sells it and the development team that builds
it once it's been sold. The result is that systems evolve slowly, and must be
built for the general case, with little understanding of the actual workloads
that run on them."
There are many more I
could have chosen. ...
read the article
- re efficiency
Warfield says - "Efficiency is important. As a rough approximation, a
server in your datacenter costs as much to power and cool over 3 years as it
does to buy up front. It is important to get every ounce of utility that you
can out of it while it is in production."
WORM hard drives - now a reality
Editor:- August 20,
2014 - From time to time I get an email from a new (to me) company which
really grabs my attention. Here's one such which arrived this morning.
now have the WORM hard disk you refer to in your article in (Introducing WORM Hard Disk
Drives - February 28, 2005).
"It was developed for the Department of Justice, and is now in
use, by GreenTec-USA, Inc. in conjunction with Seagate. Can we send you some
information? Would love to hear from you!" - Bob Waligunda,
VP of Sales at .
comments:- I haven't spoken to Bob yet - because of the time difference. But
here's some info I got from GreenTec's web site:-
- - "GreenTec Write Once Read Many (WORM) hard disk
drives protect your data from cyber attacks and from accidental modification or
The interesting thing for me is it shows that
innovation in the hard
drive market hasn't stopped completely. And GreenTec's 3TB (for now) WORM
drives are also available as arrays in micro cloud blocks.
WORM whitepaper (pdf) - "Organizations today have demanding needs to
ensure that their sensitive data is protected. Considerable damage could be done
if critical or sensitive files are deleted or altered either accidentally or
I had almost
forgotten about my 9 year old WORM HDD (market needs this) article. I'll update
it later with this note.
Linking this back to SSDs - there have been
several companies in recent quarters who have announced physical write-disable
switches into embedded SSDs - including:-
also:- SSD security,
Decloaking hidden segments in the enterprise
May 28, 2014 -
today published a new article -
hidden segments in the enterprise for rackmount SSDs
Who's got your keys?
Editor:- April 5, 2014 - "Think
about it" says Chandar
Venkataraman, Chief Product Officer,
- "If your service provider has access to your encryption keys, can you
really say that your data is secure?"
That's just one of the
thought provoking ideas in his new blog -
Things You Didn't Know About the Cloud
SSD enterprise software
Coho Data now shipping 2U MicroArray hybrids
March 6, 2014 - Coho
general availability of its first product - a 2U
SSD ASAP called the
(an SSDserver 4/E) - which
integrates PCIe SSDs,
hard drives and a
server into a web scale expandable unit (using an internal 52 port 10GbE fabric
switch) to implement what the company refers to as a "MicroArray"
designed with the philosophy of
Tiering Upside Down (pdf)" to deliver a base building block unit of
Editor's comments:- you may judge for
yourself the lofty scale of Coho's
ambitions by this market
soothsayer quote which they integrated in the launch press release - "By
2017, Web-scale IT will be an architectural approach found operating in 50%
of Global 2,000 enterprises."
SSD hybrid arrays,
meet Ken - and the
enterprise SSD software event horizon
Atlantis provides more evidence of the trend towards massively
improved enterprise utilization enabled by SSD-aware software
February 11, 2014 - Atlantis
that the new "In-Memory Storage Technology" release of its storage
virtualization software - called
can significantly increase enterprise utilization by enabling users to deploy
up to 5x more VMs on their existing storage.
USX faqs (pdf),
utilization and the SSD event horizon,
Permabit has shrunk data storage market by $300 million
Editor:- September 30, 2013 - Permabit today
and hard disk customers have shipped more than 1,000 arrays running its
and ) software in the past 6 months.
"We estimate that our
partners have delivered an astonishing $300 million in data efficiency savings
to their customers" said Tom Cook, CEO of Permabit
who anticipates license shipments to double in the next 6 months.
efficiency, new RAID in
SSDs, SSD software
SolidFire - as an anti-jitter service in the cloud
August 19, 2013 - SolidFire
provides the underlying
support for a new SSD
empowered cloud product
being offered by
which I learned about in a
by on .
Editor's comments:- cloud companies - like
the stars in the sky - are nearly numberless - however if you want to see a
partial list of who they are - is cluttered with the names of cloud companies - and reads
almost like a set of audited customer accounts than a technology news
forum - which can be off-putting - if like me - you're looking for
SSD content -
rather than SSD
But although I couldn't find any mention of this
particular story on my brief visit to their website this time around - I was
reminded about an interesting observation which SolidFire had
about earlier (in February 2013) regarding the performance and QoS
impacts that "Noisy Neighbors" can create in a shared storage
|Their leading theme is cloud
service providers - but this issue is also critical to almost any realistic
deployments in an enterprise context - and is the implicit reason that many
architects have preferred to isolate critical apps servers in the past - even
within their own datacenters - rather than risk mixing them all up in pools.
a cartoon (they call it an "infographic") -
SolidFire captures the essence of this performance randomizing problem - whose
solution (you guessed it) is to use more (of their) SSDs.
| See also:-
SSDs, can you trust
SSD performance benchmarks?,
AI in the cloud needs SSDs
September 28, 2012 - "Consumer
products are moving more and more towards that touch of artificial intelligence
and in particular speaking to your devices and having your voice sent off to the
cloud, recognised and analysed on good computers there and transmitted back"
Chief Scientist at
Fusion-io in the
interview / article -
deluge - the need for speed
Amazon offers explicit SSD performance in the cloud
July 19, 2012 - There are many ways SSDs can be used inside
classic cloud storage
services infrastructure:- to keep things running smoothly (even out
IOPS), reduce running
developers who explicitly want to access SSD like performance.
3 to 5 years time all enterprise storage infastucture will be solid state -
but due to economic necessities it will still be segmented into different types
by speed and function - as I described in my
SSD silos article -
even when it's all solid state.
I predict that when that happens -
AWS's marketers may choose to describe its lowest speed storage as "HDD
like" - even when it's SSD - in order to convey to customers what it's
about. It takes a long time for people to let go of old ideas. Remember
Virtual Tape Libraries?
magic carpet ride.|
|"What scares me is
when companies fall into the trap of trying to architect a single application to
work across multiple different cloud providers. I understand why engineers are
attracted to this.... Unfortunately, this effort eats into the productivity
gains that compelled the organization to the cloud in the first place."|
| Stephen Orban,
Global Head of Enterprise Strategy at Amazon Web 无限彩票app下载 in his
Myths about Hybrid Architectures Using the Cloud (March 5, 2015)|
|Nutanix says Pure's CEO
doesn't understand the disruptive magnitude of hyperconvergence |
|Editor:- December 6, 2016 - A new blog -
CEO disses Nutanix. OK, let's compare numbers - by Steve Kaplan, VP of
Client Strategy - Nutanix
starts out with the idea of comparing the revenue and customer acquisition
metrics of Nutanix to another well known company founded in the same year - Pure Storage -
whose financial reporting periods are the same. |
But the blog quickly
repositions to a market analysis of enterprise architecture generations and
segmentation preferences (both of which are often poorly understood in the
industry by senior managers in SSD companies).
Among other things
"Pure Storage's CEO, Scott Dietzen isn't the only
one at the storage manufacturer who doesn't grasp the magnitude of the
disruption Nutanix has brought to the datacenter." ...read
- the SSD heresies
- "Nowhere else in computer architecture will you get so many industry
experts disagreeing on such fundamental questions."
you trust SSD market data? - "These are the 5 reasons why things go
wrong in the SSD market data collection, interpretation, modeling and analysis
|At the time of writing this
blog in 2012 I was thinking of the web scale companies, the cloud
infrastructuralists, and the real time analytics jocks who would change retail,
advertising, intelligence and all kinds of data upcycling leveraged activities
which previously had been technically impossible to monetize because data
processing was too slow and the reach of memory-like latencies were too small.
|the big market impact of
SSD dark matter|
back at the online storage and cloud market|
by Zsolt Kerekes,
|This market has seen
many ups and downs in the past 15 years. |
The online backup market
flared most brightly at the height of the dotcom boom crazy days in the late
1990s. That convinced me to create a dedicated page for this subject. You can
see an archived copy of the online backup page circa 2000 -
Back then - I called it "Edrives & web based storage" - because "online
backup" hadn't yet become a standard term back then.
unconvinced about the business models for many of these companies - which mostly
relied on unsustainable web advertising. I'd been making my living from the
sustainable kind (of advertising) - and knew the difference.
enough - this segment of the storage market got itself a bad reputation for
vendor churn and undependability in the long term.
You can get a
flavor of how the online backup industry changed (and our web site too) in the
years which followed, by clicking these archived links:-
we're recently experienced another recession (caused by the credit crunch of
2008) and you've got to ask yourself this question...
If banks can
fail - then why should you trust ANY online backup provider with your data?
answer is - you shouldn't. Because
shown these services can disappear overnight.
But on the other hand -
there are many examples of where online backup has helped their customers
survive in the event of floods, fire etc.
A pragmatic approach - would
be to use 2 different types of offsite backup - which do not have common modes
of failure due to sharing software or geography. That's the way ahead for this