Why keep renting expensive resources, when you can just own it outright.
> Zoom selected Oracle to expand its cloud, bypassing major industry leaders Amazon Web Services, Alphabet’s Google Cloud Platform and Microsoft’s Azure Cloud.
I guess GCP isn't significant enough to be in the headline
"Alphabet’s Google Cloud Platform" is GCP.
(In my personal experience) Google deprecates products I love at a fast pace and have for many years.
For me, they went from an air of mystery like they were the "Willy Wonka's Chocolate Factory" of shops to "just another software conglomerate", and not a particularly good one at that.
I am curious about which products and what kind of support they are lacking at.
> (In my personal experience) Google deprecates products I love at a fast pace and have for many years.
Can you name a few? (going to guess reader, inbox for gmail etc)
And if they are being like any other tech company, why the extra unnecessary bullying though?
Look, I find their behavior pretty anti-consumer in some aspects too but relatively, they don't look bad.
We can talk about amp, controlling the web through blink and chromium, search, map and the list can go on.
But as I grow older, I think that's not how anything is going to change. Blaming the company is unreliable way to achieve what you hopefully want - a good competitive market with pro-consumer policies and behavior.
That won't happen with blaming google or any other company. People won't change either if you told them about AMP a million times everyday on every thread. This forum itself is an echo chamber in that sense as with other tech focused online groups.
Do you think you should participate in politics? Check up with your local authorities? And try to sneak in reforms that will probe up more competitors and improve the current monopoly?
The problem isn't that google is being anti-consumer more so that you don't have any way to reliably remove google as a dependency
[0] https://www.cnbc.com/2020/03/18/zoom-cfo-explains-how-the-co...
A google search confirms that they colocate, apparently with Equinix. This is as far as running a data center as living in an apartment is to building and managing apartment buildings.
(Fun Fact, while Amazon does own and run its own data centers in most places, they're also colocating with Equinix in some regions (though I assume their level of colocation goes beyond traditional colocation)).
Just about anyone who has significant network connectivity has a footprint in an Equinix datacenter. In the Bay Area you want to be in Equinix SV1 or SV5, at 11, and 9 Great Oaks, San Jose.
If you're there, you can order a cross connect to basically any telco you can imagine, and any other large company. You can also get on the Equinix exchange and connect to many more.
But, Equinix charges you a huge premium for this, typically 2 - 3x other providers for space and power. Also they charge about $300 per month per cross connect.
So your network backbone tends to have a POP here, and maybe you put some CDN nodes here, but you don't build out significant compute. It's too expensive.
On the cheaper, but still highish quality end you have companies like CoreSite, and I'm pretty sure AWS has an entire building leased out at the CoreSite SantaClara campus for portions of us-west-1. (Pretty sure because people are always cagey about this kind of thing.)
I also know that Oracle cloud has been well know for taking lots of retail and wholesale datacenter space from the likes of CoreSite, and Digital Reality Trust, because it was faster to get to market. This is compared to purpose build datacenters, which is what the larger players typically do.
In the case of AWS, I know they generally do a leaseback, where they contract with another company who owns the building shell, and then AWS brings in all their own equipment.
But all these players are also going to have some footprint in various retail datacenters like Equinix and CoreSite for the connectivity, and some extra capacity.
Zoom is probably doing a mix of various colocation providers, and just getting the best deal / quality for the given local market they want to have a PoP in. Seems like they are also making Oracle Cloud part of that story.
I’ve done big datacenter builds and cloud projects. The cloud projects were usually not a big win from a cost POV in my experience. Financially it’s a tax and marginal unit of capacity play.
Definitely easier to manage at a certain level, especially with variable demand or to accommodate growth.
So, which is it? Is it hard to say what it means exactly or is that Zoom isn’t making a big bet on OCI?
I can understand that you can be skeptical of a company deploying on OCI because of whatever biases you might have, but you’re making quite a leap that doesn’t add anything to the actual discourse on the topic.
If Zoom weren’t making a bet on OCI, they wouldn’t care to put this out into the media. They wouldn’t go on record and provide a quote if they were beholden to AWS or Azure. Zoom is a relatively small fish vs. AWS; they wouldn’t want to piss off their Cloud infrastructure provider for what you’re calling “not a big bet on OCI”.
The point stands that they have physical infrastructure and all their own stuff in cages around the world, not just cloud native AWS tools.
If your tech stack is not reliant on cloud services like S3 etc, you're better off with a cloud provider who can give you those sweet deals. But you'll need in house expertise to deal with big data.
Maybe they are afraid that Amazon or Microsoft with their tradition of copying competition would pose a threat? Even then, Microsoft is already competing using Microsoft Teams and if Amazon wanted to it wouldn't be hard for them at all to come up with a product.
How many here even knew they had a cloud you could use?
Unless it's way cheaper I don't see the point
Thanks for the laugh.
OCI is also growing very fast (aiming for 36 regions by the end of the year) and adding lots of products beyond basic compute and storage, particularly around cloud native (functions, managed Kubernetes, API gateway), of course database (classic, autonomous, MySQL, NoSQL) plus streaming, events, monitoring email delivery, marketplace etc. And a lot of Oracle products as PaaS (analytics, integration, blockchain etc.)
However there are some options here other than "Zoom management is incompetent."
1. As others have pointed out, Amazon Google Msft all have competitive products and Zoom doesn't want to give them any more insight than they already have.
2. Oracle finally has a set of cloud products that are on par with AWS, GCP or Azure from a cost, availability, license perspective
3. They got a great deal to help Oracle change the way they do business so they can become competitive in Cloud
I don't know if any of these are true or not, but none are implausible imo
I think this could be a smart move for a company to be the "big fish in a small pond" - even if that pond is owned by a shark with freaking laser beams....
Having experience with AWS, Azure, and OCI, I wouldn't voluntarily touch OCI with a 29½-foot pole. One of my company's software vendors uses OCI for the cloud-hosted version of their product, and we've had all sorts of random issues that make me really really really want to install the "on-prem" version on some Windows instances in our own AWS account. Maybe we're just an outlier and Zoom will have better experiences, but something tells me that given Zoom's scale the issues will only be amplified.
Given that it's an Oracle product, I can't imagine it being cheap in the long-run, either, so I seriously don't know what the value proposition is there. Sure, maybe Zoom got a good introductory price given their scale, but... eek.
It seems that people really eat the marketing that AWS/GCP/Azure are the best cloud provider for every single product.
Oracle Cloud's perception is in the gutter, and it's all press. It's very possible that they have a competitive cloud service. They were making some decent offers for talent last time I checked.
Maybe someone who actually works on Oracle cloud can jump in here and clarify.
I'd love to, and I'm sure others would as well, but I'm not sure what can be said that wouldn't run afoul of legal.
Zoom stressed security first, Oracle's executive team has a close relationship with the current administration. I wonder if this is a lobbying play in light of China centric security concerns.
Oracle isn't exactly known for its security, that would be Microsoft or Google.
Not a salespitch, but Oracle Cloud Infrastructure (OCI) is built with pretty serious enterprise-grade security in mind. Couple of resources:
[1] https://www.oracle.com/a/ocom/docs/oracle-cloud-infrastructu... [2] https://www.oracle.com/assets/oracle-inf-cloud-security-wp-3... [3] https://docs.cloud.oracle.com/en-us/iaas/Content/Security/Co... [4] https://blogs.oracle.com/cloud-infrastructure/the-four-pilla... [5] https://blogs.oracle.com/cloud-infrastructure/core-to-edge-s...
AWS cross zone traffic within the same region is not free:
https://www.lastweekinaws.com/blog/aws-cross-az-data-transfe...
and I distinctly remember getting bitten by the same kind of a bill at Google for traffic between their AZs within the same region, which makes me guess that Azure has the same business model.
If Oracle offered Zoom free traffic inside the region between different AZ's then it demolished AWS and GCP pricing.
Never buy Oracle.
Cognitive dissonance anybody?
Or do you need to give your soul to oracle?
"You should take them up on that offer."
The incentives available from other cloud providers are MASSIVE if your business has the chance to grow in coming years. They will literally buy your business for years on the bet that at some point they will make it back.
Even the few that don't, you can use something like Minio.
I wouldn't consider S3 something that requires lock in.
Don't think your early stage startup will ever have that problem?
That's exactly what a lot of other people thought 5-10 years ago, and now they're stuck in S3 which means they need to use EC2 to manipulate that data which means they may as well put it all in AWS because egress fees will eat their lunch.
Until you get locked in.
*SFU: selective forwarding unit - the node in a video conference call that receives and resends all the video streams between participants.
We argued against it in the dev team, but it wasn't the worst cloud migration I've done. The console reminds me of early days AWS as it's essentially just VPC+EC2+S3, but it was refreshing to spin up a server without a pages of config being presented to you. We took the opportunity to containerise our older sites and ran everything in their managed k8s cluster. I very rarely had to use the console for anything, which tbh is a bit of a grab-bag of managed services beyond the core cloud offering. Terraform support is there if you need to do anything serious.
Zoom is the success story of these corona-times of ours; the headline alone is marketing gold, particularly considering nobody else in the "startupsphere" will ever give Oracle this sort of visibility - or even the time of day. At a time when Oracle is trying to push an image of being startup-friendly, this is better than the alternatives (each word they tweet on "helping startups" unleashes waves of mean and snarky jokes).
Netflix had a 5-day outage caused by issues with their private DC back in the day.
IT mgmt. decided their expertise wasn't in operating DC's, SV real estate was too expensive, and doing multi-region themselves was too expensive.
AWS was picked as it was the only viable cloud offering at the time, and the decision was made to be mono-cloud until later. (Azure was used for storing backups.)
Note that AWS was never used for large-scale streaming. Either a partner CDN was used, or now their own CDN.
Source: worked at Netflix.
Amazon already has Chime.
Smart move, really. Build a relationship. Better than funding the competition.
If you aren't knowledgeable enough about cloud providers to evaluate them directly on their merits, then you look for signals. Some people might think, hey, Zoom is doing well, they are a well-known, up-and-coming company, and they chose Oracle, from which we infer that Oracle must be good choice.
Not heard of it? You aren’t alone there.
Microsoft has Skype.
I am not sure about Amazon. Someone please enlighten us.
Because Amazon, Microsoft, and Google are competitors to Zoom. Why would you host on your competitor?!
Or at least they used to - I guess lately they are strictly reliant on TSMC?
https://wccftech.com/samsung-aiming-high-apple-a13-next-year...
https://www.fool.com/investing/2020/01/23/apple-boosts-chip-...
Say it's about the money: then yes, you could in theory have a problem with that, but one way or the other you'll have a problem: if you work at zoom and someone has an xbox, uses windows or uses office you're still tied to that 'competitor'. It seems to me that the competitor is just that specific element (Teams, Skype) and not everything that happens to be close to me. But I'm no MBA and my perspective is probably not the most profitable one.
More fundamentally because business is cooperative and not like a war, but like a game with many repetitions. So working with your competitors can be mutually beneficial. Business is not a zero-sum game, it's not like two villages in the walking dead going at each other.
As for real reason, that’s a head scratcher...
> Zoom already uses Amazon and Microsoft’s cloud services, but went with Oracle for its latest expansion.
Remember that Oracle’s Capex spending on cloud is very small compared to AWS, Azure and GCP. Given Zoom’s growth it’s not likely they’ll choose an also-ran in the public cloud market as their sole supplier.
Increasing optionality and maybe some access to Oracle marketing is likely a bigger reason.
Yup, Larry Ellison himself recently endorsed Zoom, most likely as means to close the deal: https://youtu.be/u96GRtxBUUQ
Now they are in the news with this story ... that's valuable.
For Amazon marketplace merchants, I get how Amazon using internal metrics to start carrying popular products is a problem. For Zoom, though? They're public, and they announce quarterly usage numbers. There's nothing to copy that they can't already see.
The brand name "Zoom" is easily worth $100M on it's own now.
The current Gen 2 cloud (OCI) is very much capable. I would strongly suggest you either have a high level look at the products available (start here [1]), peruse the documentation [2], or have a go yourself with a free account [3].
Happy to continue this chat about actual products you thought were lacking, or your experience.
[1] https://www.oracle.com/cloud/ [2] https://docs.cloud.oracle.com/en-us/ [3] https://www.oracle.com/cloud/free
There's a free tier [1] which gives you an initial $300 credit for the first 30 days, then if you don't convert to a pay as you go model (absolutely your choice) will drop you to an "always free" tier that gives you a bunch of resources at no cost (2 small VMs, 2 autonomous DBs, 100GB of storage etc.)
For committed spending there is a universal credits model (which comes with a 30% discount by default).
Doesn't cost your soul either - have a look at the price list [2] or run through the cost estimator [3]. Also, some resources are always free (e.g. first 10TB of egress, cloud shell, kubernetes cluster, developer cloud services etc.)
(disclaimer, I work at Oracle)
[1] https://www.oracle.com/cloud/free/ [2] https://www.oracle.com/cloud/ucpricing.html [3] https://www.oracle.com/cloud/cost-estimator.html
So yeah, that should give everyone a pretty good idea of their reliability if they are not multi-million dollar customers.
OCI is cheaper than AWS on pretty much every metric. In this particular context:
"The Reuters article helpfully points out that Zoom has 217,000 terabytes a month of traffic flowing through it. If we assume all of that is from inside of Zoom’s environment out to the internet (it absolutely isn’t, but it’s a fine worst-case data transfer scenario) and all of it is moving to Oracle now that the deal is signed (certainly not happening, but work with me here), according to public pricing that data transfer would cost, per month: $11,186,406.55 on AWS, nobody knows on Azure because the pricing calculator thinks I’m screwing with it when I put that big of a number into it, and $1,843,630 (hat tip to Jeffery Lyon on that; I moved a decimal in an earlier version of this post) on Oracle Cloud."
https://www.lastweekinaws.com/blog/why-zoom-chose-oracle-clo...
(disclaimer, I work at Oracle)
They get enough control through their contracts to make sure the hosting provider provides exactly what they need to spec.
Those mean AWS has networking gear in those locations. You order a cross connect and plug into one of their switches.
When you walk around in one of those places, you also typically see racks of AWS gear in a smallish cage with lots of hard drives. Typically a CDN pop.
When we (locally) talk about having your own hardware we use terms like 'bare metal' and 'machines' or even the specific names of the types of machines, like 'the blades' or 'the dells'. We really only use 'datacenter' when we talk about a physical location we own, with power we own, cooling we own, networking we own, and access control we own. Otherwise it's just colo.
I live and work in the Bay area. I was part of the larger org that ran Uber’s datacenter (co-located) and that’s how it was talked about.
They buy a lot of space, and they work with lots of providers, but they don't own very many sites. So I guess they don't "have a datacenter" for the purposes of this thread. No one I know in the business thinks this way in the year 2020.
Designing and operating datacenter facilities is specialized work, and it's about compliance, auditing, risk management, electrical, plumbing, hvac and other skilled trades. The datacenter industry actually has very little to do with computers, so there is a natural split between the facility and the server / network equipment it houses.
Basically all commercial datacenter providers operate as REITs, which is tax advantageous but extremely limiting in some ways. Amazon can benefit from this (with lower pricing) without dealing with it themselves.
Owning can offer some advantages, but it also means you're with that site for the long, long haul. Efficiencies of designs are always increasing, so operating in an old facility costs you money. If you built the site to your own spec, good luck exiting -- the next owner will have to do a total overhaul to get it to industry spec and get customers.
Even if you have a 10 year lease, there are always ways to get out if you want to. Especially if you're Amazon.
But they operate at a scale that is very unique.
It's also worth noting that sometimes when a company builds a datacenter in a green field situation, it may be working with a datacenter provider on that project. So the company may own it, but they're paying the provider to use their design elements and potentially to operate it.
Teams is much better, but is the weirdest product ever. AFAIK, it consists of two voip products, bundled with a thickish client for SharePoint.
Of course it's zoom I'm talking about, come on dude.
Don't leak material information or customer data tho. Which can make it hard as some of the best stuff is secret or customer related. Filters abound.
It was weird moving to twitch which was default open before this switch as the two policies were in conflicts.
I think this is better overall. And I enjoy the insights from aws folks and gcp, and the sparring that goes on.
I don't know anyone who went to Oracle cloud and was happy, but apparently they can't share their opinions
But yea, I don't get it either. I'd be curious to know why.
Buying a chip from someone (i.e., purchasing rights to use an existing chip design made by someone) vs. ordering your own chip to be manufactured by someone are two different things.
It's like, we don't say that Apple just bought phones from Foxconn, they simply acted as a manufacturer.
However, as a customer dependent on that particular vendor, one wants to make sure that their cloud provider has no perverse incentive to deprioritize issues one encounters.
I have known people who tried to setup a data center in India and it took them around 2 years to have the first rack installed. Biggest hurdle was to get a license to store fuel in large tanks for their generators. Not to mention many of those permissions have to be renewed annually and if you fail to renew it which can take months, you are not in compliance and hence can't use the generators.
In India you can not start your own power generation plant and you can sell electricity only to the government. Depending on many situations you have to technically register a separate entity, get licenses as a "power company" then on paper sell the electricity to government and then buy it back from government for your own use.
The only actual reason to use AWS is to not divert any energy to doing anything else but scaling the company. The only problem was that by the time you are at some reasonable scale, AWS has you pretty locked in.
The fundamental problem is that AWS (or any major cloud) charges you for the amount of “stuff” you put through the pipe ($/gb), but with colocation you can pay a fixed cost for the size of the pipe ($/gbps). This allows you to do your own traffic shaping and absorb bandwidth costs without needing to pass them onto your customers.
This is the dirty, open secret of cloud pricing models. It’s also their moat, which makes it infeasible to do something like “build AWS on AWS.”
For context, if you were to buy 10Gbps of dedicated internet transit, he.net is currently advertising that for $900/month.
If we convert that to GB per month it's 3,240,000GB, so we can calculate what AWS would charge based on list prices.
Using their pricing calculator: https://calculator.aws/#/createCalculator
Outbound from Cloud Front or US West (Oregon) to the Internet:
$165,891.11
That's a 184 x increase in price!
So yeah, you have to buy networking gear and other stuff, but you can get quite a bit of gear for $165K/month. Now you don't really want to run that 10Gbps link flat out like that, but you get the point.
The AWS markup on bandwidth costs is absolutely insane.
Pro tip: if you have a large enough cloud provider spend, you can negotiate the bandwidth prices down quite a lot, given their markup, they have some room to move.
But the trick is you have to actually use it and need it in real time. An AWS instance costs you nothing if you don't use it, and almost nothing if you let them kill it at their whim.
Zoom's strategy looks pretty optimal to me. Take the 100 fold price reduction on your predicable load, farm the rest out to the lowest bidder.
Also for that amount of data it'd be faster to transfer as a stationwagon full of backup tapes (https://aws.amazon.com/snowball).
That still means to exit AWS you have to add 2.5 months of opex to your budget, which will weigh against the perceived benefits of relocating the service out of AWS. That isn't an accident on Amazon's part -- there is a business case why ingress to S3 is free and egress from S3 is not free.
There are also a lot of problems around paying both the old and new provider during the move, so if you aren't very careful you could wind up going from a $2.1 million monthly opex to $9.2 million until the project wraps up. And what projects end on time?
Snowball may speed up your timeframe and reduce costs, but 100PB still costs $3 million to move and you still need to go back and sweep up the data from S3 once you've verified the data is complete.
People forget that when you talk about customers spending 100+ mil/year they get heavy discounts on many of the SKUs.
Egress fees on the order of 40-50% off list are possible in my experience.
Dx list with 100 PiB is 2 million. Add in discounts and you’re at much less than 1 mil total
Here is an article.
https://www.nytimes.com/2019/12/15/technology/amazon-aws-clo...
The well-known US CDNs were used for streaming.
Also, the cloud they picked doesn't matter because:
1) They used their own ASG UI, Asgard, since AWS initially didn't have one.
2) With around 1,000 engineers, they could migrate clouds at any time at a future date (or build their own CDN. There was no silo politics as in most other companies.)
Just to be clear - colocation here is in order of thousands of square feet. The datacenter provider provides redundant utilities. The customer does everything else.
If someone says 'I own my house' and they actually mean 'I rent an apartment', no one is going to say 'oh well, english is an imprecise language'.
In all the examples given earlier, the words "I have a datacentre" (or the approximate "I have my own datacenter") vs "I own a datacentre" (or the approximate "I own my own datacentre") have very significant differences.
In the UK, many people would say "I have a flat in this district" and that would be understood as being a renter, since some UK neighborhoods are just too expensive, and renting is the norm rather than ownership.
Similar in North American cities "I have an apartment in a brownstone in Manhattan" - does not necessarily mean they own it, but it could mean ownership or rental or lease or rent-controlled/stabilized tenancy (which could go on for multiple decades/generations of inhabitancy).
Also, it's gotten much much better in the last 2 years. By the time I left, it's video was handily better than any non-zoom video chat I've tried, and competitive with zoom.
I use Chime and love it. Its lack of features is a feature in my opinion. I used Slack for years before using Chime and I’m happy with the switch. I don’t get nearly the chat fatigue I used to get from Slack. Chime reminds me of HipChat and Campfire in terms of UX but with the availability you’d expect an AWS service to have.
It mostly just feels like a byproduct of not-built-here syndrome.
I now find my self yearning for the stability of Chime outside of using Zoom. and using Zoom, feel a concern for my privacy in a way I have never using an AWS product.
Chime's UX design really allows you to join calls without video and without the pressure.
It's great as video doesn't add much to a large conference call.
I never had such problems with any other hosting company. I'm live between different countries so I registered it anyway, but still...
Also I found instances becoming unavailable randomly for no apparent reason so I keep my testing VMs on $5 Hetzner instances.
And yeah no way to run own mail server either because editing of reverse DNS records is not allowed on always-free tier.
If someone says "I have a brownstone in Manhattan" and they actually mean they rent one of seven rooms, it's just a lie.
OFC I'll try to do it again, but I wouldn't be posting on HN in first place if support would give anything like a useful reply.
For this client the only AWS product we were cleared to use was Lambda, and only for integrating Alexis into the product.
Your random SaaS company is often a big shitshow. I've had more than one vendor Sales Engineer show me live customer data in response to performance or other questions. Startups and smaller SaaS companies in particular often demonstrate amazing levels of cluelessness.