* Posts by cpreston

18 publicly visible posts • joined 8 Jul 2009

Uber apologises for Sydney siege surge pricing SNAFU

cpreston

it's called supply and demand

I agree with the first half of dan1980's post. If you don't like surge pricing, you don't "get" Uber. It starts with the fact that they don't schedule drivers and they are free to work or not work any time they want simply by turning on their phone and logging in. That means that at any point, demand will outstrip supply. They apply standard laws of supply and demand by temporarily raising the price, which both lowers demand by discouraging cheaper riders and increases the supply by encouraging drivers to drive on over.

If you don't get this, then don't use Uber, but this is where I differ from dan1980.

First off, Uber black car drivers are licensed drivers. Only UberX is amateurs. But it's still my and your decision to use or not use the system. But don't use it -- and the wonderful service it provides -- and then complain about how it works, which is completely different than how a taxi company works.

I for one haven't taken a taxi since I discovered Uber. I'd rather walk.

'Database failure ate my data' – Salesforce customer

cpreston

Re: An hour of transactions lost? That might be a disaster

We have a web app that talks directly to salesforce. While an hour of data would not mean millions of dollars, it would stink.

But our app is designed to handle outages and a loss of some data on the salesforce's end. We would be able to replay what happened with our web app over the last hour. It's called designing for failure.

When AWS had their big outage, there were customers who were down for a really long time -- and there were customers that experienced no downtime whatsoever. Why? Because they designed for failure. They had their systems running in multiple AWS zones and everything just moved when the one zone went down.

People that think the cloud will solve all IT problems are nuts. People that think that the cloud is crap because it doesn't solve all IT problems are also nuts. Just saying.

cpreston

I'm a salesforce customer and I can tell you that there are no SLAs. None. Seriously. So I'd say that based on that, they handled this outage pretty well.

And if you need to restore your salesforce data because YOU messed it up? That costs a minimum of $10K and takes weeks.

cpreston

"Ah the joys of CDP, now you can have two copies of that corrupted database for a very competitive price."

That's not how CDP works. True CDP can recover from any rolling disaster, including corruption. It does not appear that they were using CDP, or they would not have suffered this loss. They were apparently using near-CDP (AKA snapshots and replication), and had to go back to a snapshot from an hour ago.

cpreston

Re: Hmmm...

It's all about the SLA. If you have an SLA that promises zero data loss, then sure. it's reasonable to expect zero data loss. if you don't, then... I'd have to agree that an hour of lost data is better than most companies would do with their own data.

Microsoft secure Azure Storage goes down WORLDWIDE

cpreston

Re: This doesn't kill "the cloud" for me

@Dave Dowell

I'm sorry you find me deplorable. I'm also sorry that you misunderstood my comment about hiring IT people. I am IT people with 20 years of experience in IT. I also now employ another IT person. But we choose to use our IT skills in different ways other than supporting the day to day IT operations of our small company (<10 employees). We have "outsourced" that to various other companies. I'm sorry if you think that means that I deserve to fail. I disagree.

My choices were a lot like other small companies:

1. Hire an IT generalist that will know a little bit about everything

2. Hire a bunch of specialists via cloud companies who know a lot about a few things

I chose #2, and I don't see how that makes me an idiot. I love how you think that people who are employed by your company are inherently less risky than cloud vendors paid by your company. It's a bias that I'm trying to expose.

Please tell me why having someone on my payroll makes them less likely to screw up than someone who works for a vendor that does nothing but the thing I hired them to do (e.g. run mail services for me).

A previous commenter mentioned about the lack of control you have in a cloud situation. That is true and it's not true. It's true in that you are one customer of thousands or millions, and you have little control over how they do IT. It's not true in that if you don't like how one cloud vendor does things, there are lots of other cloud vendors to take their place.

cpreston

This doesn't kill "the cloud" for me

After commenting on a few other comments, I thought I'd make my own. First, I don't work for MS or any cloud vendor, so spare me the ad hominem attacks.

What I am is a huge user of cloud technology. In fact, my entire company runs in the cloud. Our phone system, our file sharing system, our CRM system, our payroll and accounting system, etc are all cloud providers.

Am I without outages? Absolutely not. Am I without problems? Nope. Have I fired some cloud vendors over the last few years? Yep.

But...

Do I have to hire and manage an IT staff? Nope. When I found out I didn't like a product, was I stuck with it because I paid for it up front and have to wait for it to depreciate? Nope. I just fired that cloud vendor and got another. Do I have to do any planning when I expand and contract the services I use? Nope.

I am far happier with my company's IT services than I was when I ran my last company and we did it all ourselves. And when there's a problem? I just call the guy. (2 1/2 men reference if that wasn't obvious.)

This is possibly the dumbest reason a cloud vendor has ever had an outage. But I have to quote the Big Man, "Let you who is without sin cast the first stone." I love all the comments that say they'll never trust the cloud because of things like this. As if you'd never have an outage if you did your own IT. Riiiiiigggghhht.

cpreston

Re: .. and nothing of value was lost.

I think your facts are a bit off there. The danger systems were not running on Azure, or even on updated MS equipment. And it appears that it was an outage that was reported as a data loss story. I have seen no evidence that anyone actually lost data after the restore was completed.

cpreston

Re: There are so many reasons...

1. If you think you have privacy of your contacts & email within your company, I think you'd be surprised with the reality.

2. You know absolutely nothing about what "the cloud" offers as security, do you? I've seen some of the offerings and was more impressed with that they offered than what I see in typical shops.

3. If your ISP is down, your business is down. Cloud or no cloud. But to your greater point of reliablity, I again say that just because something is in your datacenter doesn't make it more reliable.

Amen on encryption. But I have to disagree that it's the only thing the cloud is good for.

cpreston
Thumb Down

Re: Oh Rly?

Yes, because having all of your IT under your direct control always guarantees that you will never have an outage. That's why no company in the world ever reported an outage until cloud computing came around.

But I probably shouldn't bother you with facts. You're mind is obviously already made up.

Tesla vs Media again as Model S craps out on journo - on the highway

cpreston
Facepalm

The writer's story matches what Elon Musk said

Elon Musk said that the NYT writer left w/o a full charge. The writer did that twice in his story. Check. (In addition, he didn't plug it in overnight in a cold climate where he should have known he would lose some of his charge.) He said the driver drove in excess of the speed limit. Everyone knows that "keeping up with traffic" is code for speeding. Check. He said the writer took an unexpected trip in heavy traffic. The writer said he took "a short break in Manhattan.." Check.

THOSE are the facts of the story and they match what Elon Musk said. You can do one or these things, but you can't do ALL of those things and not expect a reduction in range.

cpreston
Thumb Down

He did not "follow their advice"

Here is what he said he did after he went to the last charging station:

"I should have bought a membership to Butch’s and spent a few hours there while the car charged. The displayed range never reached the number of miles remaining to Milford, and as I limped along at about 45 miles per hour I saw increasingly dire dashboard warnings to recharge immediately. "

The displayed range never reached the number of miles to Milford.

let me state that again

The displayed range never reached the number of miles to Milford.

...And so he decided to drive to Milford -- with not enough range to get there.

He may have followed their advice on how to maximize range during that final leg. But nothing was going to change the fact that he knowingly left on his final leg without enough charge for said leg.

Speaking in Tech: HP skunks Autonomy - Really, Meg... really?

cpreston
Thumb Up

Dec 3 in History

I honor of Greg's joining the dark side and moving to HDS on Dec 3, 2012, I took a look at that day in history.

2011 USA - Herman Cain Announces Suspension of Presidential Campaign

1965 USA - Ku Klux Klansmen Convicted of Murder

1967 South Africa - First Heart Transplant

But I think this one is the most apropos:

1989 Malta - George Bush and Mikhail Gorbachev declare an end to the Cold War

Here's the crack about Infosmack

cpreston
FAIL

It's a VE-GET-ABLE

What's Arugula?

it's a VE-GET-ABLE!

Watch "My Blue Heaven" starring Steve Martin. He's a Mafia boss that is put into the witness protection program and lives in San Diego. He's in the grocery store and the store manager asks him if there's anything he can get for him. He says...

Arugula!

What's that?

It's a VE-GET-ABLE.

Sainsbury's is abandoning tape

cpreston

3 PB is not fiction

Why would you think that 3 PB a night is fiction? As a consultant, I have personally worked with multiple companies that back up that kind of data. They usually have 100s of large backup servers, and thousands of tape drives to make all that happen.

EMC plus Data Domain equals what exactly?

cpreston
Stop

What?

DD does NOT do post-process. Where did you get that idea?

BTW, Dell does NOT license DXi. They are only reselling EMC's version of DXi.

Symantec eliminates dedupe disparities

cpreston
Thumb Up

Nope, you got it right

The "deduplicated archive" functionality referred to in the articles refers only to object-level deduplication (ala CAS, SIS, etc.). It's not "true" dedupe.

Yes, NBU will continue to store data in its original format on tape. I would reword your statement slightly to say that it only applies to disk targets, not just VTLs.

AFAIK, only CommVault is storing deduplicated data on tape, but I haven't been particularly excited about that idea.