5 ms·

> this is a good move by Red Hat to position themselves with rkt Red Hat does not give a damn about rkt. They care about Kubernetes, specifically establishing

by zapita 9y ago

> this is a good move by Red Hat to position themselves with rkt Red Hat does not give a damn about rkt. They care about Kubernetes, specifically establishing and consolidating a dominant position in upstream kubernetes development. CoreOS helps them do that. rkt on the other hand is a dead project and completely useless to Red Hat. > Google, on the other hand, doesn't need Docker to plug any holes in their go to market unless it's a blocking play. Google has a massive hole in their go to market. They have no clue how to sell to enterprises, and because of that they are falling behind AWS and Microsoft in the cloud market. However you're right that it would take much more than a Docker to plug that hole. Honestly if I were Google I would try to buy Red Hat, or an even larger enterprise vendor like SAP or Oracle. Not only would they be buying the enterprise distribution they lack, but they would also secure a major customer for their own infrastructure and keep up with the economies of scale AMZN and MSFT are building up. But I don't know that they will develop the humility to admit that they need that - and even if they do there's no guarantee that the integration would work... Culturally they are worlds apart. EDIT: another interesting acquisition target for Google Cloud would be Vmware. Diane Greene founded it, then got fired by EMC post-acquisition, something I am betting she still holds a grudge over. Now she runs Google Cloud. If I'm her, I like the idea of buying the place and savoring a triumphal Steve Jobs takeover moment. I don't know to what extent this is complicated by this week's announcement of a Vmware/Dell reverse merger. Maybe they buy the whole thing? But then would Michael Dell want to run this new mega-Cloud group? And how does that play with Diane's ambitions? This might bool down to whether Diane likes the idea of working for Michael Dell. He has vastly more go-to-market experience which is what matters most in their position. I really think there should be an HBO Silicon Valley - Enterprise spinoff :-)

r0fls 9y ago
I think Google being able to buy Oracle is a bit of a stretch at this point. Oracle is 1/4 the market cap of Google. So, it seems unlikely they would have the capital available.
- zapita 9y ago
  You are right. I was extrapolating future scenarios where Oracle or SAP trip up in their race to the Cloud, causing a major drop in their stock. Honestly it's a pretty realistic scenario, it's not a secret that the Dell, HPE, SAP and Oracles of the world have serious challenges ahead of them. One of them is bound to hit an attractive price, and for a bold and deep-pocketed suitor to act on it. The question is when?
  - manishjhawar 9y ago
    And Who?
kakwa_ 9y ago
Frankly, I'm not sure why Google would acquire RedHat, SAP or Oracle. I'm not sure what will be gain from that. Google probably doesn't care about the softwares these companies provides as it has its own internal alternatives/derivative and Google is not really a software vendor, so it's a bit outside of their core business. What would make more sense in my opinion would be for Google to have a partnership with a big Enterprise player (or several in fact) and promote it loudly.
- zapita 9y ago
  > Frankly, I'm not sure why Google would acquire RedHat, SAP or Oracle. Enterprise distribution. Without significant enterprise revenue, Google Cloud will never catch up to AWS and Azure. Nobody can beat Google on technology, but technology alone cannot solve this. > I'm not sure what will be gain from that. Google probably doesn't care about the softwares these companies provides as it has its own internal alternatives/derivative and Google is not really a software vendor, so it's a bit outside of their core business. Google Cloud absolutely is a software vendor. They sell cloud services which compete directly with traditional enterprise software. Their services are superior in many ways, but enterprise adoption is definitely not one of them. That is the problem they could solve by buying an enterprise-savvy competitor. As a bonus, these traditional software vendors are all racing to deploy cloud versions of their software. Because they have massive install bases, they can drive enterprise workloads to the cloud of their choice. What would make more sense in my opinion would be for Google to have a partnership with a big Enterprise player (or several in fact) and promote it loudly. They are already partnering aggressively, just like their competitors. I think they announced exactly this with SAP recently. But partnerships have their limits. Most vendors will not allow themselves to grow too dependent on their main competitors. If they did, they would lose all leverage in the deal, and slowly starve as Big Cloud grabs all the margins. And then we're back to the acquisition scenario.
- sah2ed 9y ago
  > What would make more sense in my opinion would be for Google to have a partnership with a big Enterprise player (or several in fact) and promote it loudly. They do: they have a Cisco partnership for Google Cloud. https://cloud.google.com/cisco/ https://cloud.google.com/cisco/
lima 9y ago
Red Hat cares a lot about getting rid of Docker. They're a major contributor to CRI-O [1], which is very similar to rkt. [1]: http://cri-o.io/ http://cri-o.io/
timlyo 9y ago
>rkt on the other hand is a dead project Why is this? I've only recently started looking at it.
le-mark 9y ago
I really think there should be an HBO Silicon Valley - Enterprise spinoff :-) Interesting thought, although these merger/acquisitions play out over quarters and years. Not really conducive to a 12 episode season every year. And seems Silicon Valley has already had a touch of this type of narrative :)
nailer 9y ago
> Red Hat does not give a damn about rkt. They care about Kubernetes, specifically establishing and consolidating a dominant position in upstream kubernetes development. Yep. Even technical people are used to referring to Linux container tech as 'Docker' so when you ask them what technology they use, they'll mention Kubernetes.
AndyNemmity 9y ago
Google already partners with SAP to bring larger enterprise customers to their cloud.
parasubvert 9y ago
> They have no clue how to sell to enterprises, and because of that they are falling behind AWS and Microsoft in the cloud market. One could argue the same about AWS for a long time. They’re still lumpy depending on the region. Google has gotten way better in this regard, and are hiring like mad. I wouldn’t presume the above is true any longer. Microsoft on the other hand can sell, but Azure is flaky. Google’s main limitation is they don’t have the breadth of offerings of AWS or Azure. That will change with time...
- zapita 9y ago
  > One could argue the same about AWS for a long time Yes, and it took them years to get it right. Which means Google is years behind, and doesn't have the benefit of a forgiving early market. > Google has gotten way better in this regard, and are hiring like mad. I wouldn’t presume the above is true any longer. Let's hope so for their sake. Personally I'll believe it when I see it. You can hire all the sales reps you want... If your top leadership doesn't understand or respect enterprise buyers, or know how to scale the entreprise distribution "machine", then it won't matter. I've spoken to several enterprise IT buyers - commanding billions of dollars in aggregate budget - comparing the benefits of AWS and Azure. When I asked them "what about Google Cloud", the room burst out laughing. This happened several times.
  - parasubvert 9y ago
    > Let's hope so for their sake. Personally I'll believe it when I see it. You can hire all the sales reps you want... If your top leadership doesn't understand or respect enterprise buyers, or know how to scale the entreprise distribution "machine", then it won't matter. Diane Greene knows the enterprise... she has hired an experienced team on the product, sales, and marketing side. > I've spoken to several enterprise IT buyers - commanding billions of dollars in aggregate budget - comparing the benefits of AWS and Azure. When I asked them "what about Google Cloud", the room burst out laughing. This happened several times. I too can play the anecdote game - I have senior relationships with at least a few major international banks with aggregate budgets in the billions, and they are betting on GCP. I also know a very large bank betting on AWS, and a couple are straddling the fence with multi-cloud between their on-prem, Azure, and GCP. I see Azure commitments in a big way in insurance. Of course, it depends what you're trying to do as to which cloud is the better fit. We saved a pile of money moving our AWS workloads to GCP and got better performance for it. I think laughing at Google is a mistake, but clearly they have work to do.
zaphar 9y ago
rkt on the other hand is a dead project and completely useless to Red Hat. What support do you have for this statement? The github project is still pretty active. The architecture appears to be far more sane. k8s supports it as well as docker for the runtime piece. And just anecdotally docker hasn't been the greatest experience for me running in production. So much so that I've considered exploring rkt as a replacement. I would not at all be surprised if RedHat was hedging their bets re: docker with this coreos acquisition.
- zapita 9y ago
  I'm sure rkt is well engineered (overall CoreOS has good open-source engineering). It just doesn't have much usage, and whatever advantages it has, evidently it wasn't enough to convince a meaningful number of Docker users to switch. I use Docker intensively and, like you, I've had my share of frustrating issues with it over the years. But at the end of the day, the bugs get fixed reasonably quickly, and Docker just gets the job done. Their code has so much mileage at this point, I just don't see the point in starting over with a competitor. Evidently I'm not the only one who feels that way.
  - zaphar 9y ago
    I don't know. I wouldn't say my experience has been that the bugs get fixed reasonably quickly. It's more like they make several attempts to fix the bugs reasonably quickly and then eventually figure it out. I'm not sure their code has enough mileage to paper over their quality issues. At this point the only reason I'm not switching over is that I've written enough shims and code to paper over dockers problems that I have it somewhat stable and the switch cost is just barely high enough to make sticking with it the right choice for now. But it won't take too many more problems to make me reevaluate that decision.
    - zapita 9y ago
      It's surprising that our experiences were so different. Are you using CE or EE? And if it's CE, is it the upstream-supplied package or the distro-supplied one? In my experience those things can make a big difference.
tripzilch 9y ago
Google should buy Oracle, fire all the employees[0] except Larry Ellison and have him mow their lawns. [0] or reassign to better projects

by zapita 9y ago

r0fls 9y ago

I think Google being able to buy Oracle is a bit of a stretch at this point. Oracle is 1/4 the market cap of Google. So, it seems unlikely they would have the capital available.

zapita 9y ago
You are right. I was extrapolating future scenarios where Oracle or SAP trip up in their race to the Cloud, causing a major drop in their stock. Honestly it's a pretty realistic scenario, it's not a secret that the Dell, HPE, SAP and Oracles of the world have serious challenges ahead of them. One of them is bound to hit an attractive price, and for a bold and deep-pocketed suitor to act on it. The question is when?
- manishjhawar 9y ago
  And Who?

kakwa_ 9y ago

Frankly, I'm not sure why Google would acquire RedHat, SAP or Oracle. I'm not sure what will be gain from that. Google probably doesn't care about the softwares these companies provides as it has its own internal alternatives/derivative and Google is not really a software vendor, so it's a bit outside of their core business. What would make more sense in my opinion would be for Google to have a partnership with a big Enterprise player (or several in fact) and promote it loudly.

zapita 9y ago
> Frankly, I'm not sure why Google would acquire RedHat, SAP or Oracle. Enterprise distribution. Without significant enterprise revenue, Google Cloud will never catch up to AWS and Azure. Nobody can beat Google on technology, but technology alone cannot solve this. > I'm not sure what will be gain from that. Google probably doesn't care about the softwares these companies provides as it has its own internal alternatives/derivative and Google is not really a software vendor, so it's a bit outside of their core business. Google Cloud absolutely is a software vendor. They sell cloud services which compete directly with traditional enterprise software. Their services are superior in many ways, but enterprise adoption is definitely not one of them. That is the problem they could solve by buying an enterprise-savvy competitor. As a bonus, these traditional software vendors are all racing to deploy cloud versions of their software. Because they have massive install bases, they can drive enterprise workloads to the cloud of their choice. What would make more sense in my opinion would be for Google to have a partnership with a big Enterprise player (or several in fact) and promote it loudly. They are already partnering aggressively, just like their competitors. I think they announced exactly this with SAP recently. But partnerships have their limits. Most vendors will not allow themselves to grow too dependent on their main competitors. If they did, they would lose all leverage in the deal, and slowly starve as Big Cloud grabs all the margins. And then we're back to the acquisition scenario.
sah2ed 9y ago
> What would make more sense in my opinion would be for Google to have a partnership with a big Enterprise player (or several in fact) and promote it loudly. They do: they have a Cisco partnership for Google Cloud. https://cloud.google.com/cisco/ https://cloud.google.com/cisco/

lima 9y ago

Red Hat cares a lot about getting rid of Docker. They're a major contributor to CRI-O [1], which is very similar to rkt. [1]: http://cri-o.io/ http://cri-o.io/

timlyo 9y ago

>rkt on the other hand is a dead project Why is this? I've only recently started looking at it.

le-mark 9y ago

I really think there should be an HBO Silicon Valley - Enterprise spinoff :-) Interesting thought, although these merger/acquisitions play out over quarters and years. Not really conducive to a 12 episode season every year. And seems Silicon Valley has already had a touch of this type of narrative :)

nailer 9y ago

> Red Hat does not give a damn about rkt. They care about Kubernetes, specifically establishing and consolidating a dominant position in upstream kubernetes development. Yep. Even technical people are used to referring to Linux container tech as 'Docker' so when you ask them what technology they use, they'll mention Kubernetes.

AndyNemmity 9y ago

Google already partners with SAP to bring larger enterprise customers to their cloud.

parasubvert 9y ago

> They have no clue how to sell to enterprises, and because of that they are falling behind AWS and Microsoft in the cloud market. One could argue the same about AWS for a long time. They’re still lumpy depending on the region. Google has gotten way better in this regard, and are hiring like mad. I wouldn’t presume the above is true any longer. Microsoft on the other hand can sell, but Azure is flaky. Google’s main limitation is they don’t have the breadth of offerings of AWS or Azure. That will change with time...

zapita 9y ago
> One could argue the same about AWS for a long time Yes, and it took them years to get it right. Which means Google is years behind, and doesn't have the benefit of a forgiving early market. > Google has gotten way better in this regard, and are hiring like mad. I wouldn’t presume the above is true any longer. Let's hope so for their sake. Personally I'll believe it when I see it. You can hire all the sales reps you want... If your top leadership doesn't understand or respect enterprise buyers, or know how to scale the entreprise distribution "machine", then it won't matter. I've spoken to several enterprise IT buyers - commanding billions of dollars in aggregate budget - comparing the benefits of AWS and Azure. When I asked them "what about Google Cloud", the room burst out laughing. This happened several times.
- parasubvert 9y ago
  > Let's hope so for their sake. Personally I'll believe it when I see it. You can hire all the sales reps you want... If your top leadership doesn't understand or respect enterprise buyers, or know how to scale the entreprise distribution "machine", then it won't matter. Diane Greene knows the enterprise... she has hired an experienced team on the product, sales, and marketing side. > I've spoken to several enterprise IT buyers - commanding billions of dollars in aggregate budget - comparing the benefits of AWS and Azure. When I asked them "what about Google Cloud", the room burst out laughing. This happened several times. I too can play the anecdote game - I have senior relationships with at least a few major international banks with aggregate budgets in the billions, and they are betting on GCP. I also know a very large bank betting on AWS, and a couple are straddling the fence with multi-cloud between their on-prem, Azure, and GCP. I see Azure commitments in a big way in insurance. Of course, it depends what you're trying to do as to which cloud is the better fit. We saved a pile of money moving our AWS workloads to GCP and got better performance for it. I think laughing at Google is a mistake, but clearly they have work to do.

zaphar 9y ago

rkt on the other hand is a dead project and completely useless to Red Hat. What support do you have for this statement? The github project is still pretty active. The architecture appears to be far more sane. k8s supports it as well as docker for the runtime piece. And just anecdotally docker hasn't been the greatest experience for me running in production. So much so that I've considered exploring rkt as a replacement. I would not at all be surprised if RedHat was hedging their bets re: docker with this coreos acquisition.

zapita 9y ago
I'm sure rkt is well engineered (overall CoreOS has good open-source engineering). It just doesn't have much usage, and whatever advantages it has, evidently it wasn't enough to convince a meaningful number of Docker users to switch. I use Docker intensively and, like you, I've had my share of frustrating issues with it over the years. But at the end of the day, the bugs get fixed reasonably quickly, and Docker just gets the job done. Their code has so much mileage at this point, I just don't see the point in starting over with a competitor. Evidently I'm not the only one who feels that way.
- zaphar 9y ago
  I don't know. I wouldn't say my experience has been that the bugs get fixed reasonably quickly. It's more like they make several attempts to fix the bugs reasonably quickly and then eventually figure it out. I'm not sure their code has enough mileage to paper over their quality issues. At this point the only reason I'm not switching over is that I've written enough shims and code to paper over dockers problems that I have it somewhat stable and the switch cost is just barely high enough to make sticking with it the right choice for now. But it won't take too many more problems to make me reevaluate that decision.
  - zapita 9y ago
    It's surprising that our experiences were so different. Are you using CE or EE? And if it's CE, is it the upstream-supplied package or the distro-supplied one? In my experience those things can make a big difference.

tripzilch 9y ago

Google should buy Oracle, fire all the employees[0] except Larry Ellison and have him mow their lawns. [0] or reassign to better projects