Today, Arq 5.9 was released.
Adds support for two new providers: Backblaze B2 and Wasabi.
Naturally, I asked "Who the hell are Wasabi?"... Here's the long answer by digging around online...
Answer:
The founders of Carbonite (online backup solution) have left Carbonite to start a cloud storage company. They were building the company in secret and unveiled in May of 2017.
Right now they have 1 data center and 20 employees. And they have $8.5 million in venture funding from 3 investors:
https://******************/organization/wasabi-technologies-inc#/entity
I looked at the investors hoping to see if anyone would give the impression of being able to properly judge the success chances of the company, but none of them give any "we understand cloud storage" impression, so not much to learn there:
https://******************/person/ron-skates#/entity
https://******************/person/howard-cox#/entity
https://******************/person/desh-deshpande#/entity
Beyond the $8.5, they also raised +$10.7 million recently, so that's a total of ~$20 million in their funds:
https://www.americaninno.com/boston...m-round-carbonite-founders-raises-for-wasabi/
And the CEO commented (in the link above) that he had to stop the 2nd fundraiser because they had more money than they needed (wtf?) and that he may decide to add $10 million of his own money next year (wtf?).
So at least they're very well-funded... People are throwing money at these ex-Carbonite guys...
The link above also reveals that the $8.5 million would have lasted them "to the end of the year" (may to december = 7 months)... and that they did the 2nd fundraiser just to get $1 million extra for marketing, and that they ended up with a lot more than they needed. This is all very strange!
Friend said Wasabi’s customer traction has been “fantastic.” While Wasabi was only expecting a couple dozen companies to start trials in the first couple months, it ended up attracting nearly 600 so far, “with very high conversion rates.” The need for customer support has been low, Friend added, since Wasabi works very similar to how S3 does.
That's a big "wtf?". So the founder only expected "a couple dozen companies" to start trials in the first few months. But they got over 600 in a few days. How on EARTH did he not foresee the massive interest in such a dirt-cheap S3-compatible cloud storage claiming to be as reliable as Amazon?! It's as if he's out of touch... Weird.
In their press releases, they're talking about being ready for "exabytes" of data, and having "0.99999999999 (eleven 9's)" reliability. I really don't know about that. This whole thing could be snake oil. Nobody knows at this point.
Their company twitter:
https://twitter.com/wasabi_cloud (really weird timeline full of random tech articles, as if they are trying to create discussion about random subjects to get seen).
CEO twitter:
https://twitter.com/Wasabi_Dave
May 2 CEO tweet: "Holy cow! We already had to stop new signups because we're concerned about running out of capacity." (
https://twitter.com/Wasabi_Dave/status/859775051251494912)
Another on May 2: "Will have more storage online in two weeks. Get in line! First come first served."
Interesting May 4 tweet: "Day 2 as the world's most cost effective cloud storage. Seems like we underestimated the demand by quite a bit. Good problem to have." (
https://twitter.com/Wasabi_Dave/status/860190321602424833)
Here is a video from someone on the team, speaking about his confidence in the Carbonite founders and that they know how to build online storage:
https://twitter.com/wasabi_cloud/status/886646400775270401
The fact that they founded Carbonite and have experience with huge data centers and cloud storage gives me some hope that this is not just a "so cheap and underpriced that it's gonna crash and burn" company.
I also hope that their hardware infrastructure is good and won't crash and burn under the load... But it seems like they're already getting customers faster than they can expand storage capacity?!
Lastly, I worry that they are just sleazily subsidizing storage costs via the fundraised money and that they are just cheap right now to get tons of customers and publicity, and that they will do a price hike later. It's inevitable that they will need to raise this price... But they'll probably still be really cheap if that happens...
Here is a brief company presentation in the words of the CEO:
https://medium.com/@wasabi_cloud/welcome-to-wasabi-hot-storage-4e06d58e377c
Today Wasabi will launch cloud storage that is so fast, so cheap and so reliable that it will mark the beginning of cloud storage as a commodity. Wasabi is the hot storage company — fast to write, fast to read and instantly available. Instead of industry behemoths doing their best to lock customers into high-priced proprietary storage, Wasabi’s cloud storage is open, easy to use and 100% compatible with the Amazon S3 API. No vendor lock-in. No need for silly artificial storage tiers: Wasabi is 6X the speed and 1/5 the price of S3, even cheaper than Glacier. And more reliable than either. Built for exabyte scale. We’re the same team that created Carbonite, one of the world’s largest cloud backup companies. Welcome to the next generation of cloud storage. Welcome to Wasabi hot storage!
David Friend, Co-Founder and CEO, Wasabi
Here is an article which reveals some good stuff:
http://www.crn.com/news/storage/300...-market-with-disruptive-price-performance.htm
From that article, we learn that David Friend co-founded Carbonite, was its CEO for 10 years, and left it in January of 2015 to found Wasabi. Which was in stealth mode until May of 2017. That's a lot of time out of his life in "stealth mode" building this thing, which makes me think it's serious... Hmm.
And here's some super interesting quotes from that article:
Friend declined to discuss the technology behind the Wasabi cloud storage offering, other than to say that is not the same technology used at Carbonite. Wasabi and Carbonite do not have any equity relationship, he said.
"The cost of storage depends on the cost of hard drives," he said. "We get the same price as the other cloud providers. Our secret sauce is in the software. We focus on how to manage the data. We don't use Windows or Linux. We have learned in the last 15 years how to control that cost. Our technology is completely new."
Wasabi will initially work mainly direct with end-user customers who will likely sign on to the service via the web and a credit card, Friend said. "Our marketing job is to make sure people know us," he said. "People who care about the cost of storage will choose Wasabi. People who care about blinding speed will choose Wasabi."
So that's fascinating. They do not use stock Windows or Linux. They use something custom-written to store data. Perhaps that's how they are able to be "6x faster than Amazon S3"? Perhaps they've programmed some FPGAs to do networking and storage management. I have no idea... they are very secretive about it.
It's actually worrying that they are not using stock Windows or Linux... because that means they've most likely custom-written, and that means bugs (I'm an exceptionally skilled programmer and hell yes there are bugs in ALL software, especially while it's fresh and new).
The article also goes on to add some comments from some tech guy in the business of online storage, who is a bit skeptical about how well they'll work out:
If Wasabi's service works as stated, it will be a very interesting offering, said Rafi Kronzon, CEO of Cartwheel, a New York-based managed service provider.
However, Kronzon told CRN, it will be important to see how well it actually works.
"The actual speed depends on many things," he said. "Speed moves at the bottleneck rate. Where's the bottleneck? In the upload? The download? The processing of the data?"
MSPs, in general, do not compete head-to-head with public cloud storage providers like Amazon, Kronzon said. "We usually store data in S3, but only through backup services like Carbonite, Backupify, and Crashplan," he said. "If Wasabi is going after these vendors, or after companies like Dropbox, it would be a big deal. But Amazon has all the right tools to store and archive data."
Wasabi is entering a crowded cloud storage market that is quickly racing to a cost of zero, Kronzon said.
"There's lots of room in the market, but it's still a tough market," he said. "I would have thought there would actually be more consolidation in this market by now. Cost-per-gigabyte is not a big deal for customers with a terabyte or two of data. The big difference is at large enterprises or at reseller selling services with petabytes of data."
But best of all is this article:
http://www.pcworld.com/article/3194...new-cloud-service-like-low-priced-wasabi.html
Here are some of the most important quotes from it:
The company, started by the co-founders of online backup provider Carbonite, says its single pool of capacity can deliver primary, secondary or archive data at a sustained-read speed of 1.3GB per second, versus 191MB per second at Amazon. Its durability is the same, Wasabi says.
Still, enterprises may approach a new service like Wasabi with caution at first, Enterprise Strategy Group analyst Steve Duplessie said.
"I can see companies with huge data sets that have at least two copies in the cloud moving the secondary to a much cheaper alternative like Wasabi," Duplessie said.
Price is just one factor enterprises should consider when they consider a new cloud storage service.
Also, take a good look at whether the service will deliver the performance you need, Duplessie said. And with startups, it's worth considering a company's business practices and financial footing.
"Are they going to be there tomorrow? Are they going to be able to continue supporting you?" Connor said.
Wasabi knows it will need to prove itself and that risk-averse companies may start with second copies of data, CEO David Friend said. "Over time, people will grow to trust us."
The company isn't saying much about what makes the service fast or cheap. Part of it involves the way its software breaks up data and reads it off multiple hard drives in parallel. There is some flash in the architecture, too.
Wasabi has just one data center, located in Northern Virginia and linked to AWS over Amazon's Direct Connect system. Customers can load data into Amazon's EC2 compute cloud faster from Wasabi than from S3, Friend said. Wasabi plans to build another data center in a different location and will charge customers to replicate their data to that site. There are also plans for overseas facilities eventually.
The service isn't designed for small or short-lived data, like temporary data on a website, Friend said. It's for companies with hundreds of terabytes or more that they're going to keep for at least two weeks.
Wasabi offers mostly the same features as S3, with one addition, called "immutable buckets," that the company says will be a boon to data integrity. These are data sets that can't be changed by anyone for a set period of time, determined by the customer's business policies. Immutable buckets could be used for data such as X-rays that a medical center needs to keep for a long time, Friend said.
So from those quotes, the most important things are: It can apparently restore data very, very fast (much faster than Amazon S3). They are using some sort of RAID (multiple harddrives in parallel for data safety and speed). And they are planning a second datacenter in USA and then expansion overseas (Europe!). And anytime a new datacenter opens up, they will let customers move their data there instead (for a fee).
So, with all of that backstory out of the way... here's how Backblaze B2 and Wasabi compare against each other:
Backblaze B2:
- Storage cost per gigabyte: $0.005/GB.
- Download cost per gigabyte: $0.02/GB. But you get 1 GB of free downloads per day, which is great for doing occasional, minor file restores for free.
- There is no minimum monthly charge. You only pay for what you use.
- There is also a very minor API call cost, but it's so small that you won't notice it.
- Incredibly good storage architecture, which has fully proven its reliability. And they have recently expanded to having 2 data centers.
- The company is reliable and won't vanish or hike its prices.
- Perfect storage reliability. They're splitting data across something like 20 hard disks, and each disk is in a separate storage "pod" (cabinet), and out of those they're using 17+3 parity which means that they can lose 3 whole cabinets/3 whole disks out of the set and can still recover the data (and they recover data transparently). The whole system is incredibly well-engineered.
Wasabi:
- Storage cost per gigabyte: $0.0039/GB. (-22% of the cost of B2 storage, so about 4/5ths of the B2 cost).
- Download cost per gigabyte: $0.04/GB. (+100% of the B2 cost... 2x as expensive as B2).
- Edit: I just found out that if you store less than 1 TB (1024 GB), they'll charge you as if you had 1 TB. Meaning their minimum monthly charge is actually $3.99. So some of the comparison numbers below are wrong. But I won't re-calculate them since they're still correct when you disregard the minimum charge.
- There is no cost for API calls.
- But there is a PRETTY BIG caveat with their billing system: Every object (file chunk) you upload is pre-billed for 90 days of storage (and that amount is non-refundable even if you delete the file again immediately).
- We don't know if they'll do a big price hike soon. They are marketing themselves as an Amazon S3 API-compatible competitor to Amazon, which is ~$0.023/GB, so Wasabi has a LONG way that they can raise their $0.0039/GB and still be totally competitive against Amazon. And since they're not competing against B2 (different markets; Wasabi is S3-compatible and B2 isn't), then they may raise their price far above B2's cost. I am almost certain that Wasabi is doing an aggressive company-launch marketing campaign and are planning to get a lot of customers while living off of fundraised money, and then doing a huuuuge price hike (which at that point will still be cheaper than S3), and hoping companies and customers who needed their S3-compatibility stay on their "still cheaper than Amazon" service.
- Further proof of that theory is that they said that the $8.5 million would last them for 7 months which shows they're running a very expensive, losing operation right now...
- We don't even know if they'll stay afloat. But it does seem like they're competent enough to stay alive. And of course their goal is to make a successful company to earn them money.
Storage price comparisons for 700 GB:
- B2 Monthly Cost: 700 * 0.005 = $3.5
- Wasabi Monthly Cost: 700 * 0.0039 = $2.73.
- B2 Total Yearly: $42.
- Wasabi Total Yearly: $32.76.
- B2 Download cost (700 GB): $14.
- Wasabi Download cost (700 GB): $28.
Storage price comparisons for 1300 GB:
- B2 Monthly Cost: 1300 * 0.005 = $6.5.
- Wasabi Monthly Cost: 1300 * 0.0039 = $5.07.
- B2 Total Yearly: $78.
- Wasabi Total Yearly: $60.84.
- B2 Download cost (1300 GB): $26.
- Wasabi Download cost (1300 GB): $52.
So... on paper, for a normal home user's backup set at such small scales... there's hardly any difference in the yearly storage costs. And as for their difference in download cost, you never really need to download anything via Arq unless your hard drives and other backup systems all die, so the download cost is no huge reason to choose one over the other.
But the thing is... I am almost certain that Wasabi is going to raise their insanely low STORAGE price. $0.0039/GB is not sustainable. That's $11.7 per month to store 3000 gigabytes, which is enough to fill up a 3 TB hard drive, which costs about $83 each... So you'd need to store that data for 7 months for them to make a profit off of the cost of just the hard drive itself. Add in the costs of employees, hardware, electricity, etc, and you're looking at a lot longer time before they make a profit, if ever... Hmm...
Remember that Wasabi offers an Amazon S3 API and competes against Amazon's $0.023/GB cost. So let's say Wasabi raises the price to $0.02. That's still less than Amazon and therefore companies who need S3 API (a lot do) would stay with Wasabi, but suddenly it's 4x the cost of Backblaze B2 (which does not support S3 API).
In short: Who the hell knows what to do... Wasabi could be anything... Their founder has cloud storage experience from Carbonite and has worked in secret since January of 2015 to set up this new company. But they could die in a year or two due to the cut-throat "race to zero dollar" cloud storage market and their unsustainably low prices, or they could thrive and expand globally. They could die under the load of too many customers (see their early tweets a day after launch), or they could be super fast. They could raise the price to far beyond B2's price, and the risk of them doing that is pretty high after they've gained enough customers... Or, they could keep their $0.0039 and forever be 4/5ths the price of B2 storage...
I had signed up with a B2 trial in preparation for Arq's B2 support... and I think I'll still choose B2...
There are so many risks with Wasabi. Sure, I'd pay $60.84 ($17.16 less than B2) per year instead of $78 to store 1300 GB... but, then they'll probably raise the price and suddenly it'll be something like $90 per year for Wasabi. Do I really want the hassle of switching provider AGAIN if Wasabi, an unproven company with a "try to make a splash with low fees" launch marketing campaign, suddenly raises their prices? They're competing with S3 (and their API); NOT with B2. So they are free to raise their price much higher than B2, and it's almost inevitable that they'll do so... In fact, Wasabi could raise their current prices to far above B2's price and they
still wouldn't piss off people who come from Amazon S3's massive prices.
So, B2 is a much safer choice. For just $17 more per year, about the cost of a single pizza and coke at a restaurant, I can choose B2 instead and rest safe knowing that Backblaze B2 is a trusted company with a proven track record and a stable price. And an incredibly safe 17+3 hard disk parity storage architecture which will never lose a single byte of my data (
https://www.backblaze.com/blog/vault-cloud-storage-architecture/).
You will all have to ask yourselves the same question. I don't know Wasabi's future. Time will tell if Wasabi is trustworthy. They may have a truly badly designed system, and they may be unsustainable and massively hike their prices as soon as they've ensnared enough customers.
I can say for sure: Switching providers sucks. It drains time and energy. I had about 1300 GB at Amazon Cloud Drive. It took about 2 weeks to upload all of that data. Then I had to just delete it all, which loses all history records... And I will now have to spend weeks doing the migration to the new provider, during which time I have no online backups... I don't know if it's worth trying Wasabi just because it's
currently 4/5th's (~80%) of the B2 storage cost (while being twice as expensive at downloads)...
I am not interested in losing time and energy to be a tester for an unproven company (Wasabi), just to save a measly $17 per year ($1.4 per month) while risking a sudden, massive price hike or a collapse of the company... Heck, I buy candy bars that cost more than that damn $1.4 price difference per month.
So I'm going to choose B2. Affordable,
completely proven as a company, and super reliable storage. Let's go!