This author really doesn't understand IBM's systems business.
IBM STG (hardware) is much more important than you make it out to be. It doesn't appear that way on the earnings report because many components of the STG business are software and services revenue. These are systems, not hardware. For instance, purchase an IBM Power-AIX server and IBM accounts for the AIX, PowerHA, PowerVM, etc licenses as part of software group. Likewise in the storage business (DS8 software stack, XIV software stack, etc). Even more so with mainframe where you not only have the base OS stack, but CICS, VSAM, MQ, DB2 for z, etc, etc which only run on z. Same situation with System i, which is predominantly software. If you account for all the software which is tied to STG systems (i.e. included software stack), you are talking about a large and profitable business.
As a corollary to that point, IBM's independent software (IM, WebSphere, Tivoli, etc) are generally sold with the IBM systems. For instance, Tivoli Storage Manager and Tivoli Storage Productivity Center are sold with IBM storage. Not many EMC users purchase EMC storage and then decide to purchase Tivoli for storage management functions. They purchase Avamar and the EMC stack instead. If you get rid of Power, you are going to take a hit in DB2 and the rest of IM. If you sell off storage, you are going to take a hit in Tivoli.
On the services side, it is a similar picture. All services attached to IBM hardware are part of GTS (Global Services). For instance, if you purchase a DS8 and four years of maintenance and support. All of that support value is Global Services revenue. Though, obviously, if you didn't have a DS8, you wouldn't be purchasing DS8 support.
Taken in total, hardware (z, Power and storage) is a much larger business than represented on the earnings because a large portion of the revenue, and almost all of the profit, is software or services. NetApp's hardware business, for instance, would look pretty terrible too if you considered all the NetApp software/services as completely unrelated to the hardware and only looked at the P&L on the bare metal.
Having written all of that, IBM's storage is down as compared to last year and the year before for the reasons you wrote, but apparently didn't appreciate. IBM's former disk storage business had three major components DS8, NetApp rebrand (FAS - N-Series) and Engenio rebrand. IBM is no longer selling the second two pillars, so, naturally, there is going to be a transition period from OEMed to internally developed storage systems (XIV, V7000 and FlashSystem). Eventually it will be to the positive as this is all IBM IP and are generally solid systems. All three of those internally developed storage systems, XIV, V7000, and Flash, are growing by double digits.
XIV is a "niche" product with "little or no go forward." The EMC t-shirt is in the mail.... XIV has grown rapidly every year of its existence. There are many thousands of XIV full frames in the field many of which replaced EMC Symm. Hardly a "niche." It bothers me that analysts, so-called, are still just taking EMC's word for it on XIV. Even if you agree with EMC on the XIV architecture, you can't deny there is a ton of it in the field and it isn't a blip on the radar screen.