Oracle Said to Be Leading Anti-Amazon Lobby on Pentagon Cloud Bid
bloomberg.com
bloomberg.com
I moved a lot of infra out of AWS and into GKE which I'm loving. Outside of GKE, which is great. GCP has a few things that I like better than AWS.
I also moved some infra into Azure, specifically into ACS, and then AKS and now ACS-engine.
It is very much not great.
I exaggerate but not by much.
Don't get me wrong, not a huge fan of Oracle but am curious to know if there is an actual problem, or if this is simply a hater-ade/fanboy party with no substance.
On the other hand, without the problems, the solution engineers might not exist...
...which is, in one sentence, the reason why Oracle consultants exist.
You've never had to deal with Oracle sales or "support" staff or -- even worse -- manage or support Oracle products in production, have you? If you had, well, you would already know the answer. Oracle RDBMS might be the one exception.
(FWIW, I mean "real" Oracle products, not, say, MySQL or Oracle Linux, for example.)
Ignore the comments and just try it. Run an instance for a weekend, try to setup a site or install a DB and see how it goes. I think you'll learn very quickly why the reputation exists.
VM.Standard.1.1 and VM.Standard1.16, same cost but the latter has 16 cores instead of 1 (same CPU), 112 GB memory instead of 7.
The only thing it has in common with a proper cloud thing like AWS or GCP is the 'cloud' in the name.
The rest of their products are managed services running on this, similar to the other clouds, and they work fine for what they are. Bandwidth costs are always the limiting factor though with any cross-cloud situation and their Watson AI is useless and nowhere near what they hype.
Could you elaborate on why this is a good thing?
- Support is so hard to reach that you need support to get support. Requires creating a separate user account just to file a ticket and I've since been unable to log back in because the Oracle SSO was somehow not connected to our tenant's identity instance (which I cant find in our console), but I cant get them to reset since I must file tickets from that account. There seem to be 3 different documentation sites and they link to PDF books.
- There are only 4 regions globally, 2 in the US. No concept of availability zones. Starting an instance may take minutes or hours, and they have outdated images for anything not Oracle Linux. In past 2 weeks, there were 3 emergency maintenance events. Maintenance is not automatic and there is no concept of "live migration" or any attempt to not reboot your VM. Networking is nowhere near listed capacity in use.
- Managed services are completely separated from IaaS resources. They can take hours to deploy. Less control (as expected) but still require maintenance packs to be applied manually. Maintenance can also take hours. Event hubs service doesn't even show you how much disk is available. Seems like they are nothing more than templates to trigger some instances in a hidden cloud account.
- There is no pricing within the console so you must reference documentation. This doesn't cover any hidden pricing for operations. There is no billing dashboard anywhere, so you don't know costs at all until the bill comes.
- Too tired to list anything else. There is no advantage compared to any other cloud or even 2nd tier colo. The prices are also more expensive. It is a nightmare we are forced to use for the integration cloud since Oracle has no modern concept of "apps" for any of its products.
As a result, their sales and business practices focus on selling huge deployments to Big Enterprise and large government contracts.
Not surprising they never really made inroads with small developers and startups, and why they have the reputation they do in that community.
Their business applications on the other hand I have never used but have heard they are pretty much horrible.
But now Oracle is going to companies and saying hey you don’t need DBAs, use our cloud. The right thing for all Oracle DBAs to do right now is to push for their orgs to move wholescale to Postgres. Hit Oracle where it hurts.
How can you make sure you have a better deal... when the process is not open to you? The article says open Amazon are being allowed to bid and the first aim of Oracle is to just be allowed to bid.
> Their goal is to make sure that the award process is opened up to more than one company and unseat Amazon as the front-runner for the multibillion-dollar deal.
> The Pentagon has said it intends to move the department’s technology needs -- 3.4 million users and 4 million devices -- to the cloud, indicating the massive size of the award. Pentagon officials, including Defense Secretary Jim Mattis, have repeatedly said no decision about the winner-take-all contract has been pre-made and that bids will be considered on their merits, with an award to a company or a team of companies expected in September.
What rational reason would there be for a corporation or government entity to spread its information across multiple cloud products? That would be like mandating that half a company use LibreOffice and the other half Office365 for the sake of "equality".
You might not like data center management, but the trade you are making is giving hypervisor control to a third party with 10s of thousands of employees outside of your oversight which also happens to be one of the most valuable targets for hackers in the world. You could easily have AWS employees (or hackers who have compromised the control network) dumping everything interesting from your HD images to the highest bidder and you have no means to detect this.
AWS is okay if you are only collecting relatively innocuous consumer account info, but it's completely unacceptable for any companies holding data/executing processes that can have a major impact on society when it leaks/fails.
Oracle is quite bad. Procurement in the Government is broken. The DOD is the worst. It's literally a give away to the private sector. If you think that someone "wins" this kind of business because their "product" "is the best"...You are going to be very sad.
Very few and weak managed service offerings, APIs are mostly in beta, SDKs and libraries are in alpha, support is overly sensitive, often wrong, and will take days to reply unless you select P1 priority.
They are the best if you only need strong IaaS or run on GKE, otherwise AWS is literally turn-key to run your business with every imaginable product available.
I don't know what to say buddy. I wish it mattered. It doesn't. It just doesn't. They will all get a slice of the pie. They all do the lawsuit and contract BS dance. Some of them are better at PR than others. Some of them currently, and in the past, have made so much money, and have so captured the segment of government that they sell to, that they couldn't give a rat's a what you think. Oh, and you can't do anything about it. So, maybe feeling emotionally invested in who get's to run the virtual machines is one way of dealing with it.
The only fix for this is more participation in government...like I said...I'm a pessimist.
I would trust an arbitrary product which competes on the marketplace and is sold into government on a COTS basis more than virtually anything arising out of government/vendor insanity. (I'd probably go with a purely-Government solution if it were a really specialized thing and of small enough scale, developed organically by the end user org, more than a traditional procurement thing, but COTS generally is the winner whenever a COTS option exists.)
"Trust a product"...that's part of the problem right there. Why do we have to trust them? It's just weird how emotion becomes part of this. I'm not trying to single you out or anything. I do it too, reflexively.
I guess my point is that the only difference between Oracle and Amazon are some letters in the name and some circumstances. If you think that Amazon won't behave in exactly the same way under a different set of circumstances, then you are going to be let down.
Maybe I'm just rambling...I'm pretty pessimistic about the state of the IT industry and American business in general. It's all just so incredibly short sighted.
It's the same thing when USG buys efficient, highly-optimized products like, say, regular pickup trucks polarises and such for fleets, rather than paying LockMart/NG/Boeing/BAE to build them a custom "non-tactical advanced transport device".
Idk about you, but those are probably the top 3 attributes I want in a company that manages my infrastructure.
What “manner” are you talking about?
I think fragmentation is the problem.
(Disclaimer: I've been out of the defense contracting business for some time)
What do you get out of the deal? Software that makes AWS look like...AWS.
All three vendors appear to have settled on Kubernetes as the next level of abstraction. It would be great to see dollars that would have otherwise gone to Oracle enhance the capabilities of k8s.
I never understand why the us government wants to contract these things.
Ideally, they’d contribute to OpenStack through such an initiative, ensuring an open source alternative to commercial closed cloud providers.
I don’t want my tax dollars going towards Bezo’s attempts at a retail and cloud computing monopoly, but that’s just me. Maybe there’s an opportunity here for Trump considering his disdain for Bezos and the USDS being part of the Executive Office of the President.
Nah.
I ran a couple data centers on either coast each burning about half a megawatt, which isn't that much in the grand scheme of thing. My employer at the time offered a SaaS product to Fortune 100 companies. The product required a realtime kernel and QOS in the network, neither of which is supported in AWS.
I did a fair bit of math comparing costs. Typically, you could pay off the CapEx with 6 months of usage in AWS.
IIRC we had around $8m in salaries for the ops org. I'm not sure we would've needed a great deal fewer people having all that infra in AWS.
The real problem comes from trying to find a team of people willing to engineer your service for failure. That kind of attitude just doesn't exist in the world of IT, so anyone would be hard pressed to put a team together.
Good luck finding a network engineering team that knows BGP well enough for such a thing. Good luck finding a team of engineers that know Openstack well enough to keep it running. Good luck finding folks to run Ceph. Good luck getting all that right.
If you could, though, there are certainly savings to be had.
How many data centers do you think Oracle operates in the US, presently?
Or two departments have essentially the same requirements, but they compete for budget and each have to justify their staffing by duplicating effort.
(These pathologies aren't unique to government. You can also find them in large private companies.)
I think that falls under the `other reasons`
Building data centers is expensive and on-demand compute is nice
On demand compute is nice for startups, but overly expensive for mature organizations.
You cannot afford to own a refinery. Delta Airlines can and does, as a hedge against fuel costs (which turned out to be a very profitable decision for them). That is the point.
Governments outsource because the best people in various industries aren't working for the government (for lots of different reasons).
Government IT expertise is not high-quality and will never match what AWS has. It's about value, not cost. It's better to have reliable and scalable infrastructure even if AWS is vastly more expensive.
Is there a good reason for this contract to a single cloud service provider?
AWS already runs the CIA's private cloud.
Edit: make _all_ products open source Apache license style
https://www.google.ca/amp/s/amp.reddit.com/r/Android/comment...
sounds reasonable.
An upstream connectivity outage should be nothing more than an inconvenience, not your lifeline to your lifeline to your only provider of your digital business process and records.